Datasets:

MaartenGr

/

arxiv_nlp

like 2

Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Models

Much recent work seeks to evaluate values and opinions in large language models (LLMs) using multiple-choice surveys and questionnaires. Most of this work is motivated by concerns around real-world LLM applications. For example, politically-biased LLMs may subtly influence society when they are used by millions of people. Such real-world concerns, however, stand in stark contrast to the artificiality of current evaluations: real users do not typically ask LLMs survey questions. Motivated by this discrepancy, we challenge the prevailing constrained evaluation paradigm for values and opinions in LLMs and explore more realistic unconstrained evaluations. As a case study, we focus on the popular Political Compass Test (PCT). In a systematic review, we find that most prior work using the PCT forces models to comply with the PCT's multiple-choice format. We show that models give substantively different answers when not forced; that answers change depending on how models are forced; and that answers lack paraphrase robustness. Then, we demonstrate that models give different answers yet again in a more realistic open-ended answer setting. We distill these findings into recommendations and open challenges in evaluating values and opinions in LLMs.

Computation and Language

Set the Clock: Temporal Alignment of Pretrained Language Models

Language models (LMs) are trained on web text originating from many points in time and, in general, without any explicit temporal grounding. This work investigates the temporal chaos of pretrained LMs and explores various methods to align their internal knowledge to a target time, which we call "temporal alignment." To do this, we first automatically construct a dataset containing 20K time-sensitive questions and their answers for each year from 2000 to 2023. Based on this dataset, we empirically show that pretrained LMs (e.g., LLaMa2), despite having a recent pretraining cutoff (e.g., 2022), mostly answer questions using earlier knowledge (e.g., in 2019). We then develop several methods, from prompting to finetuning, to align LMs to use their most recent knowledge when answering questions, and investigate various factors in this alignment. Our experiments show that aligning LLaMa2 to the year 2022 can boost its performance by up to 62% relatively as measured by that year, even without mentioning time information explicitly, indicating the possibility of aligning models' internal sense of time after pretraining. Finally, we find that alignment to a historical time is also possible, with up to 2.8$\times$ the performance of the unaligned LM in 2010 if finetuning models to that year. These findings hint at the sophistication of LMs' internal knowledge organization and the necessity of tuning them properly.

Computation and Language

OncoGPT: A Medical Conversational Model Tailored with Oncology Domain Expertise on a Large Language Model Meta-AI (LLaMA)

In the past year, there has been a growing trend in applying Large Language Models (LLMs) to the field of medicine, particularly with the advent of advanced language models such as ChatGPT developed by OpenAI. However, there is limited research on LLMs specifically addressing oncology-related queries. The primary aim of this research was to develop a specialized language model that demonstrates improved accuracy in providing advice related to oncology. We performed an extensive data collection of online question-answer interactions centered around oncology, sourced from reputable doctor-patient platforms. Following data cleaning and anonymization, a dataset comprising over 180K+ oncology-related conversations was established. The conversations were categorized and meticulously reviewed by field specialists and clinicians to ensure precision. Employing the LLaMA model and other selected open-source datasets, we conducted iterative fine-tuning to enhance the model's proficiency in basic medical conversation and specialized oncology knowledge. We observed a substantial enhancement in the model's understanding of genuine patient inquiries and its reliability in offering oncology-related advice through the utilization of real online question-answer interactions in the fine-tuning process. We release database and models to the research community (https://github.com/OncoGPT1).

Computation and Language

Investigating the Effectiveness of HyperTuning via Gisting

Gisting (Mu et al., 2023) is a simple method for training models to compress information into fewer token representations using a modified attention mask, and can serve as an economical approach to training Transformer-based hypernetworks. We introduce HyperLlama, a set of Gisting-based hypernetworks built on Llama-2 models that generates task-specific soft prefixes based on few-shot inputs. In experiments across P3, Super-NaturalInstructions and Symbol Tuning datasets, we show that HyperLlama models can effectively compress information from few-shot examples into soft prefixes. However, they still underperform multi-task fine-tuned language models with full attention over few-shot in-context examples. We also show that HyperLlama-generated soft prefixes can serve as better initializations for further prefix tuning. Overall, Gisting-based hypernetworks are economical and easy to implement, but have mixed empirical performance.

Computation and Language

Nemotron-4 15B Technical Report

We introduce Nemotron-4 15B, a 15-billion-parameter large multilingual language model trained on 8 trillion text tokens. Nemotron-4 15B demonstrates strong performance when assessed on English, multilingual, and coding tasks: it outperforms all existing similarly-sized open models on 4 out of 7 downstream evaluation areas and achieves competitive performance to the leading open models in the remaining ones. Specifically, Nemotron-4 15B exhibits the best multilingual capabilities of all similarly-sized models, even outperforming models over four times larger and those explicitly specialized for multilingual tasks.

Computation and Language

Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Prompts

As large language models (LLMs) become increasingly prevalent across many real-world applications, understanding and enhancing their robustness to user inputs is of paramount importance. Existing methods for identifying adversarial prompts tend to focus on specific domains, lack diversity, or require extensive human annotations. To address these limitations, we present Rainbow Teaming, a novel approach for producing a diverse collection of adversarial prompts. Rainbow Teaming casts adversarial prompt generation as a quality-diversity problem, and uses open-ended search to generate prompts that are both effective and diverse. It can uncover a model's vulnerabilities across a broad range of domains including, in this paper, safety, question answering, and cybersecurity. We also demonstrate that fine-tuning on synthetic data generated by Rainbow Teaming improves the safety of state-of-the-art LLMs without hurting their general capabilities and helpfulness, paving the path to open-ended self-improvement.

Computation and Language

A Survey on Data Selection for Language Models

A major factor in the recent success of large language models is the use of enormous and ever-growing text datasets for unsupervised pre-training. However, naively training a model on all available data may not be optimal (or feasible), as the quality of available text data can vary. Filtering out data can also decrease the carbon footprint and financial costs of training models by reducing the amount of training required. Data selection methods aim to determine which candidate data points to include in the training dataset and how to appropriately sample from the selected data points. The promise of improved data selection methods has caused the volume of research in the area to rapidly expand. However, because deep learning is mostly driven by empirical evidence and experimentation on large-scale data is expensive, few organizations have the resources for extensive data selection research. Consequently, knowledge of effective data selection practices has become concentrated within a few organizations, many of which do not openly share their findings and methodologies. To narrow this gap in knowledge, we present a comprehensive review of existing literature on data selection methods and related research areas, providing a taxonomy of existing approaches. By describing the current landscape of research, this work aims to accelerate progress in data selection by establishing an entry point for new and established researchers. Additionally, throughout this review we draw attention to noticeable holes in the literature and conclude the paper by proposing promising avenues for future research.

Computation and Language

Mysterious Projections: Multimodal LLMs Gain Domain-Specific Visual Capabilities Without Richer Cross-Modal Projections

Multimodal large language models (MLLMs) like LLaVA and GPT-4(V) enable general-purpose conversations about images with the language modality. As off-the-shelf MLLMs may have limited capabilities on images from domains like dermatology and agriculture, they must be fine-tuned to unlock domain-specific applications. The prevalent architecture of current open-source MLLMs comprises two major modules: an image-language (cross-modal) projection network and a large language model. It is desirable to understand the roles of these two modules in modeling domain-specific visual attributes to inform the design of future models and streamline the interpretability efforts on the current models. To this end, via experiments on 4 datasets and under 2 fine-tuning settings, we find that as the MLLM is fine-tuned, it indeed gains domain-specific visual capabilities, but the updates do not lead to the projection extracting relevant domain-specific visual attributes. Our results indicate that the domain-specific visual attributes are modeled by the LLM, even when only the projection is fine-tuned. Through this study, we offer a potential reinterpretation of the role of cross-modal projections in MLLM architectures. Projection webpage: https://claws-lab.github.io/projection-in-MLLMs/

Computation and Language

Eight Methods to Evaluate Robust Unlearning in LLMs

Machine unlearning can be useful for removing harmful capabilities and memorized text from large language models (LLMs), but there are not yet standardized methods for rigorously evaluating it. In this paper, we first survey techniques and limitations of existing unlearning evaluations. Second, we apply a comprehensive set of tests for the robustness and competitiveness of unlearning in the "Who's Harry Potter" (WHP) model from Eldan and Russinovich (2023). While WHP's unlearning generalizes well when evaluated with the "Familiarity" metric from Eldan and Russinovich, we find i) higher-than-baseline amounts of knowledge can reliably be extracted, ii) WHP performs on par with the original model on Harry Potter Q&A tasks, iii) it represents latent knowledge comparably to the original model, and iv) there is collateral unlearning in related domains. Overall, our results highlight the importance of comprehensive unlearning evaluation that avoids ad-hoc metrics.

Computation and Language

Do Large Language Models Latently Perform Multi-Hop Reasoning?

We study whether Large Language Models (LLMs) latently perform multi-hop reasoning with complex prompts such as "The mother of the singer of 'Superstition' is". We look for evidence of a latent reasoning pathway where an LLM (1) latently identifies "the singer of 'Superstition'" as Stevie Wonder, the bridge entity, and (2) uses its knowledge of Stevie Wonder's mother to complete the prompt. We analyze these two hops individually and consider their co-occurrence as indicative of latent multi-hop reasoning. For the first hop, we test if changing the prompt to indirectly mention the bridge entity instead of any other entity increases the LLM's internal recall of the bridge entity. For the second hop, we test if increasing this recall causes the LLM to better utilize what it knows about the bridge entity. We find strong evidence of latent multi-hop reasoning for the prompts of certain relation types, with the reasoning pathway used in more than 80% of the prompts. However, the utilization is highly contextual, varying across different types of prompts. Also, on average, the evidence for the second hop and the full multi-hop traversal is rather moderate and only substantial for the first hop. Moreover, we find a clear scaling trend with increasing model size for the first hop of reasoning but not for the second hop. Our experimental findings suggest potential challenges and opportunities for future development and applications of LLMs.

Computation and Language

MobiLlama: Towards Accurate and Lightweight Fully Transparent GPT

"Bigger the better" has been the predominant trend in recent Large Language Models (LLMs) development. However, LLMs do not suit well for scenarios that require on-device processing, energy efficiency, low memory footprint, and response efficiency. These requisites are crucial for privacy, security, and sustainable deployment. This paper explores the "less is more" paradigm by addressing the challenge of designing accurate yet efficient Small Language Models (SLMs) for resource constrained devices. Our primary contribution is the introduction of an accurate and fully transparent open-source 0.5 billion (0.5B) parameter SLM, named MobiLlama, catering to the specific needs of resource-constrained computing with an emphasis on enhanced performance with reduced resource demands. MobiLlama is a SLM design that initiates from a larger model and applies a careful parameter sharing scheme to reduce both the pre-training and the deployment cost. Our work strives to not only bridge the gap in open-source SLMs but also ensures full transparency, where complete training data pipeline, training code, model weights, and over 300 checkpoints along with evaluation codes is available at : https://github.com/mbzuai-oryx/MobiLlama.

Computation and Language

Long Dialog Summarization: An Analysis

Dialog summarization has become increasingly important in managing and comprehending large-scale conversations across various domains. This task presents unique challenges in capturing the key points, context, and nuances of multi-turn long conversations for summarization. It is worth noting that the summarization techniques may vary based on specific requirements such as in a shopping-chatbot scenario, the dialog summary helps to learn user preferences, whereas in the case of a customer call center, the summary may involve the problem attributes that a user specified, and the final resolution provided. This work emphasizes the significance of creating coherent and contextually rich summaries for effective communication in various applications. We explore current state-of-the-art approaches for long dialog summarization in different domains and benchmark metrics based evaluations show that one single model does not perform well across various areas for distinct summarization tasks.

Computation and Language

What Do Language Models Hear? Probing for Auditory Representations in Language Models

This work explores whether language models encode meaningfully grounded representations of sounds of objects. We learn a linear probe that retrieves the correct text representation of an object given a snippet of audio related to that object, where the sound representation is given by a pretrained audio model. This probe is trained via a contrastive loss that pushes the language representations and sound representations of an object to be close to one another. After training, the probe is tested on its ability to generalize to objects that were not seen during training. Across different language models and audio models, we find that the probe generalization is above chance in many cases, indicating that despite being trained only on raw text, language models encode grounded knowledge of sounds for some objects.

Computation and Language

Benchmarking LLMs on the Semantic Overlap Summarization Task

Semantic Overlap Summarization (SOS) is a constrained multi-document summarization task, where the constraint is to capture the common/overlapping information between two alternative narratives. While recent advancements in Large Language Models (LLMs) have achieved superior performance in numerous summarization tasks, a benchmarking study of the SOS task using LLMs is yet to be performed. As LLMs' responses are sensitive to slight variations in prompt design, a major challenge in conducting such a benchmarking study is to systematically explore a variety of prompts before drawing a reliable conclusion. Fortunately, very recently, the TELeR taxonomy has been proposed which can be used to design and explore various prompts for LLMs. Using this TELeR taxonomy and 15 popular LLMs, this paper comprehensively evaluates LLMs on the SOS Task, assessing their ability to summarize overlapping information from multiple alternative narratives. For evaluation, we report well-established metrics like ROUGE, BERTscore, and SEM-F1$ on two different datasets of alternative narratives. We conclude the paper by analyzing the strengths and limitations of various LLMs in terms of their capabilities in capturing overlapping information The code and datasets used to conduct this study are available at https://anonymous.4open.science/r/llm_eval-E16D.

Computation and Language

Can Large Language Models Recall Reference Location Like Humans?

When completing knowledge-intensive tasks, humans sometimes need not just an answer but also a corresponding reference passage for auxiliary reading. Previous methods required obtaining pre-segmented article chunks through additional retrieval models. This paper explores leveraging the parameterized knowledge stored during the pre-training phase of large language models (LLMs) to independently recall reference passage from any starting position. We propose a two-stage framework that simulates the scenario of humans recalling easily forgotten references. Initially, the LLM is prompted to recall document title identifiers to obtain a coarse-grained document set. Then, based on the acquired coarse-grained document set, it recalls fine-grained passage. In the two-stage recall process, we use constrained decoding to ensure that content outside of the stored documents is not generated. To increase speed, we only recall a short prefix in the second stage, then locate its position to retrieve a complete passage. Experiments on KILT knowledge-sensitive tasks have verified that LLMs can independently recall reference passage location in various task forms, and the obtained reference significantly assist downstream tasks.

Computation and Language

DiffuCOMET: Contextual Commonsense Knowledge Diffusion

Inferring contextually-relevant and diverse commonsense to understand narratives remains challenging for knowledge models. In this work, we develop a series of knowledge models, DiffuCOMET, that leverage diffusion to learn to reconstruct the implicit semantic connections between narrative contexts and relevant commonsense knowledge. Across multiple diffusion steps, our method progressively refines a representation of commonsense facts that is anchored to a narrative, producing contextually-relevant and diverse commonsense inferences for an input context. To evaluate DiffuCOMET, we introduce new metrics for commonsense inference that more closely measure knowledge diversity and contextual relevance. Our results on two different benchmarks, ComFact and WebNLG+, show that knowledge generated by DiffuCOMET achieves a better trade-off between commonsense diversity, contextual relevance and alignment to known gold references, compared to baseline knowledge models.

Computation and Language

Towards Explainability and Fairness in Swiss Judgement Prediction: Benchmarking on a Multilingual Dataset

The assessment of explainability in Legal Judgement Prediction (LJP) systems is of paramount importance in building trustworthy and transparent systems, particularly considering the reliance of these systems on factors that may lack legal relevance or involve sensitive attributes. This study delves into the realm of explainability and fairness in LJP models, utilizing Swiss Judgement Prediction (SJP), the only available multilingual LJP dataset. We curate a comprehensive collection of rationales that `support' and `oppose' judgement from legal experts for 108 cases in German, French, and Italian. By employing an occlusion-based explainability approach, we evaluate the explainability performance of state-of-the-art monolingual and multilingual BERT-based LJP models, as well as models developed with techniques such as data augmentation and cross-lingual transfer, which demonstrated prediction performance improvement. Notably, our findings reveal that improved prediction performance does not necessarily correspond to enhanced explainability performance, underscoring the significance of evaluating models from an explainability perspective. Additionally, we introduce a novel evaluation framework, Lower Court Insertion (LCI), which allows us to quantify the influence of lower court information on model predictions, exposing current models' biases.

Computation and Language

Z-AGI Labs at ClimateActivism 2024: Stance and Hate Event Detection on Social Media

In the digital realm, rich data serves as a crucial source of insights into the complexities of social, political, and economic landscapes. Addressing the growing need for high-quality information on events and the imperative to combat hate speech, this research led to the establishment of the Shared Task on Climate Activism Stance and Hate Event Detection at CASE 2024. Focused on climate activists contending with hate speech on social media, our study contributes to hate speech identification from tweets. Analyzing three sub-tasks - Hate Speech Detection (Sub-task A), Targets of Hate Speech Identification (Sub-task B), and Stance Detection (Sub-task C) - Team Z-AGI Labs evaluated various models, including LSTM, Xgboost, and LGBM based on Tf-Idf. Results unveiled intriguing variations, with Catboost excelling in Subtask-B (F1: 0.5604) and Subtask-C (F1: 0.7081), while LGBM emerged as the top-performing model for Subtask-A (F1: 0.8684). This research provides valuable insights into the suitability of classical machine learning models for climate hate speech and stance detection, aiding informed model selection for robust mechanisms.

Computation and Language

Multi-Task Contrastive Learning for 8192-Token Bilingual Text Embeddings

We introduce a novel suite of state-of-the-art bilingual text embedding models that are designed to support English and another target language. These models are capable of processing lengthy text inputs with up to 8192 tokens, making them highly versatile for a range of natural language processing tasks such as text retrieval, clustering, and semantic textual similarity (STS) calculations. By focusing on bilingual models and introducing a unique multi-task learning objective, we have significantly improved the model performance on STS tasks, which outperforms the capabilities of existing multilingual models in both target language understanding and cross-lingual evaluation tasks. Moreover, our bilingual models are more efficient, requiring fewer parameters and less memory due to their smaller vocabulary needs. Furthermore, we have expanded the Massive Text Embedding Benchmark (MTEB) to include benchmarks for German and Spanish embedding models. This integration aims to stimulate further research and advancement in text embedding technologies for these languages.

Computation and Language

Leveraging Large Language Models for Learning Complex Legal Concepts through Storytelling

Making legal knowledge accessible to non-experts is crucial for enhancing general legal literacy and encouraging civic participation in democracy. However, legal documents are often challenging to understand for people without legal backgrounds. In this paper, we present a novel application of large language models (LLMs) in legal education to help non-experts learn intricate legal concepts through storytelling, an effective pedagogical tool in conveying complex and abstract concepts. We also introduce a new dataset LegalStories, which consists of 295 complex legal doctrines, each accompanied by a story and a set of multiple-choice questions generated by LLMs. To construct the dataset, we experiment with various LLMs to generate legal stories explaining these concepts. Furthermore, we use an expert-in-the-loop method to iteratively design multiple-choice questions. Then, we evaluate the effectiveness of storytelling with LLMs through an RCT experiment with legal novices on 10 samples from the dataset. We find that LLM-generated stories enhance comprehension of legal concepts and interest in law among non-native speakers compared to only definitions. Moreover, stories consistently help participants relate legal concepts to their lives. Finally, we find that learning with stories shows a higher retention rate for non-native speakers in the follow-up assessment. Our work has strong implications for using LLMs in promoting teaching and learning in the legal field and beyond.

Computation and Language

Re-Ex: Revising after Explanation Reduces the Factual Errors in LLM Responses

Mitigating hallucination issues is one of the main challenges of LLMs we need to overcome, in order to reliably use them in real-world scenarios. Recently, various methods are proposed to check the factual errors in the LLM-generated texts and revise them accordingly, to reduce the hallucination issue. In this paper, we propose Re-Ex, a method of revising LLM-generated texts, which introduces a novel step dubbed as the factual error explanation step. Re-Ex revises the initial response of LLMs using 3-steps: first, external tools are used to get the evidences on the factual errors in the response; second, LLMs are instructed to explain the problematic parts of the response based on the evidences gathered in the first step; finally, LLMs revise the response using the explanation obtained in the second step. In addition to the explanation step, we propose new prompting techniques to reduce the amount of tokens and wall-clock time required for the response revision process. Compared with existing methods including Factool, CoVE, and RARR, Re-Ex provides better revision performance with less time and fewer tokens in multiple benchmarks.

Computation and Language

Creating Suspenseful Stories: Iterative Planning with Large Language Models

Automated story generation has been one of the long-standing challenges in NLP. Among all dimensions of stories, suspense is very common in human-written stories but relatively under-explored in AI-generated stories. While recent advances in large language models (LLMs) have greatly promoted language generation in general, state-of-the-art LLMs are still unreliable when it comes to suspenseful story generation. We propose a novel iterative-prompting-based planning method that is grounded in two theoretical foundations of story suspense from cognitive psychology and narratology. This theory-grounded method works in a fully zero-shot manner and does not rely on any supervised story corpora. To the best of our knowledge, this paper is the first attempt at suspenseful story generation with LLMs. Extensive human evaluations of the generated suspenseful stories demonstrate the effectiveness of our method.

Computation and Language

Fact-and-Reflection (FaR) Improves Confidence Calibration of Large Language Models

For a LLM to be trustworthy, its confidence level should be well-calibrated with its actual performance. While it is now common sense that LLM performances are greatly impacted by prompts, the confidence calibration in prompting LLMs has yet to be thoroughly explored. In this paper, we explore how different prompting strategies influence LLM confidence calibration and how it could be improved. We conduct extensive experiments on six prompting methods in the question-answering context and we observe that, while these methods help improve the expected LLM calibration, they also trigger LLMs to be over-confident when responding to some instances. Inspired by human cognition, we propose Fact-and-Reflection (FaR) prompting, which improves the LLM calibration in two steps. First, FaR elicits the known "facts" that are relevant to the input prompt from the LLM. And then it asks the model to "reflect" over them to generate the final answer. Experiments show that FaR prompting achieves significantly better calibration; it lowers the Expected Calibration Error by 23.5% on our multi-purpose QA tasks. Notably, FaR prompting even elicits the capability of verbally expressing concerns in less confident scenarios, which helps trigger retrieval augmentation for solving these harder instances.

Computation and Language

Clustering Document Parts: Detecting and Characterizing Influence Campaigns From Documents

We propose a novel clustering pipeline to detect and characterize influence campaigns from documents. This approach clusters parts of document, detects clusters that likely reflect an influence campaign, and then identifies documents linked to an influence campaign via their association with the high-influence clusters. Our approach outperforms both the direct document-level classification and the direct document-level clustering approach in predicting if a document is part of an influence campaign. We propose various novel techniques to enhance our pipeline, including using an existing event factuality prediction system to obtain document parts, and aggregating multiple clustering experiments to improve the performance of both cluster and document classification. Classifying documents on the top of clustering not only accurately extracts the parts of the documents that are relevant to influence campaigns, but also capture influence campaigns as a coordinated and holistic phenomenon. Our approach makes possible more fine-grained and interpretable characterizations of influence campaigns from documents.

Computation and Language

Extreme Encoder Output Frame Rate Reduction: Improving Computational Latencies of Large End-to-End Models

The accuracy of end-to-end (E2E) automatic speech recognition (ASR) models continues to improve as they are scaled to larger sizes, with some now reaching billions of parameters. Widespread deployment and adoption of these models, however, requires computationally efficient strategies for decoding. In the present work, we study one such strategy: applying multiple frame reduction layers in the encoder to compress encoder outputs into a small number of output frames. While similar techniques have been investigated in previous work, we achieve dramatically more reduction than has previously been demonstrated through the use of multiple funnel reduction layers. Through ablations, we study the impact of various architectural choices in the encoder to identify the most effective strategies. We demonstrate that we can generate one encoder output frame for every 2.56 sec of input speech, without significantly affecting word error rate on a large-scale voice search task, while improving encoder and decoder latencies by 48% and 92% respectively, relative to a strong but computationally expensive baseline.

Computation and Language

An Effective Mixture-Of-Experts Approach For Code-Switching Speech Recognition Leveraging Encoder Disentanglement

With the massive developments of end-to-end (E2E) neural networks, recent years have witnessed unprecedented breakthroughs in automatic speech recognition (ASR). However, the codeswitching phenomenon remains a major obstacle that hinders ASR from perfection, as the lack of labeled data and the variations between languages often lead to degradation of ASR performance. In this paper, we focus exclusively on improving the acoustic encoder of E2E ASR to tackle the challenge caused by the codeswitching phenomenon. Our main contributions are threefold: First, we introduce a novel disentanglement loss to enable the lower-layer of the encoder to capture inter-lingual acoustic information while mitigating linguistic confusion at the higher-layer of the encoder. Second, through comprehensive experiments, we verify that our proposed method outperforms the prior-art methods using pretrained dual-encoders, meanwhile having access only to the codeswitching corpus and consuming half of the parameterization. Third, the apparent differentiation of the encoders' output features also corroborates the complementarity between the disentanglement loss and the mixture-of-experts (MoE) architecture.

Computation and Language

When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Method

While large language models (LLMs) often adopt finetuning to unlock their capabilities for downstream applications, our understanding on the inductive biases (especially the scaling properties) of different finetuning methods is still limited. To fill this gap, we conduct systematic experiments studying whether and how different scaling factors, including LLM model size, pretraining data size, new finetuning parameter size and finetuning data size, affect the finetuning performance. We consider two types of finetuning -- full-model tuning (FMT) and parameter efficient tuning (PET, including prompt tuning and LoRA), and explore their scaling behaviors in the data-limited regime where the LLM model size substantially outweighs the finetuning data size. Based on two sets of pretrained bilingual LLMs from 1B to 16B and experiments on bilingual machine translation and multilingual summarization benchmarks, we find that 1) LLM finetuning follows a powerbased multiplicative joint scaling law between finetuning data size and each other scaling factor; 2) LLM finetuning benefits more from LLM model scaling than pretraining data scaling, and PET parameter scaling is generally ineffective; and 3) the optimal finetuning method is highly task- and finetuning data-dependent. We hope our findings could shed light on understanding, selecting and developing LLM finetuning methods.

Computation and Language

Measuring Vision-Language STEM Skills of Neural Models

We introduce a new challenge to test the STEM skills of neural models. The problems in the real world often require solutions, combining knowledge from STEM (science, technology, engineering, and math). Unlike existing datasets, our dataset requires the understanding of multimodal vision-language information of STEM. Our dataset features one of the largest and most comprehensive datasets for the challenge. It includes 448 skills and 1,073,146 questions spanning all STEM subjects. Compared to existing datasets that often focus on examining expert-level ability, our dataset includes fundamental skills and questions designed based on the K-12 curriculum. We also add state-of-the-art foundation models such as CLIP and GPT-3.5-Turbo to our benchmark. Results show that the recent model advances only help master a very limited number of lower grade-level skills (2.5% in the third grade) in our dataset. In fact, these models are still well below (averaging 54.7%) the performance of elementary students, not to mention near expert-level performance. To understand and increase the performance on our dataset, we teach the models on a training split of our dataset. Even though we observe improved performance, the model performance remains relatively low compared to average elementary students. To solve STEM problems, we will need novel algorithmic innovations from the community.

Computation and Language

Reasoning in Conversation: Solving Subjective Tasks through Dialogue Simulation for Large Language Models

Large Language Models (LLMs) have achieved remarkable performance in objective tasks such as open-domain question answering and mathematical reasoning, which can often be solved through recalling learned factual knowledge or chain-of-thought style reasoning. However, we find that the performance of LLMs in subjective tasks is still unsatisfactory, such as metaphor recognition, dark humor detection, etc. Compared to objective tasks, subjective tasks focus more on interpretation or emotional response rather than a universally accepted reasoning pathway. Based on the characteristics of the tasks and the strong dialogue-generation capabilities of LLMs, we propose RiC (Reasoning in Conversation), a method that focuses on solving subjective tasks through dialogue simulation. The motivation of RiC is to mine useful contextual information by simulating dialogues instead of supplying chain-of-thought style rationales, thereby offering potential useful knowledge behind dialogues for giving the final answers. We evaluate both API-based and open-source LLMs including GPT-4, ChatGPT, and OpenChat across twelve tasks. Experimental results show that RiC can yield significant improvement compared with various baselines.

Computation and Language

MATHSENSEI: A Tool-Augmented Large Language Model for Mathematical Reasoning

Tool-augmented Large Language Models (TALM) are known to enhance the skillset of large language models (LLM), thereby, leading to their improved reasoning abilities across many tasks. While, TALMs have been successfully employed in different question-answering benchmarks, their efficacy on complex mathematical reasoning benchmarks, and the potential complimentary benefits offered by tools for knowledge retrieval and mathematical equation solving, are open research questions. In this work, we present MATHSENSEI, a tool-augmented large language model for mathematical reasoning. Augmented with tools for knowledge retrieval (Bing Web Search), program execution (Python), and symbolic equation solving (Wolfram-Alpha), we study the complimentary benefits of these tools through evaluations on mathematical reasoning datasets. We perform exhaustive ablations on MATH,a popular dataset for evaluating mathematical reasoning on diverse mathematical disciplines. We also conduct experiments involving well-known tool planners to study the impact of tool sequencing on the model performance. MATHSENSEI achieves 13.5% better accuracy over gpt-3.5-turbo with chain-of-thought on the MATH dataset. We further observe that TALMs are not as effective for simpler math word problems (in GSM-8k), and the benefit increases as the complexity and required knowledge increases (progressively over AQuA, MMLU-Math, and higher level complex questions in MATH). The code and data are available at https://github.com/Debrup-61/MathSensei.

Computation and Language

Beyond the Known: Investigating LLMs Performance on Out-of-Domain Intent Detection

Out-of-domain (OOD) intent detection aims to examine whether the user's query falls outside the predefined domain of the system, which is crucial for the proper functioning of task-oriented dialogue (TOD) systems. Previous methods address it by fine-tuning discriminative models. Recently, some studies have been exploring the application of large language models (LLMs) represented by ChatGPT to various downstream tasks, but it is still unclear for their ability on OOD detection task.This paper conducts a comprehensive evaluation of LLMs under various experimental settings, and then outline the strengths and weaknesses of LLMs. We find that LLMs exhibit strong zero-shot and few-shot capabilities, but is still at a disadvantage compared to models fine-tuned with full resource. More deeply, through a series of additional analysis experiments, we discuss and summarize the challenges faced by LLMs and provide guidance for future work including injecting domain knowledge, strengthening knowledge transfer from IND(In-domain) to OOD, and understanding long instructions.

Computation and Language

Speak Out of Turn: Safety Vulnerability of Large Language Models in Multi-turn Dialogue

Large Language Models (LLMs) have been demonstrated to generate illegal or unethical responses, particularly when subjected to "jailbreak." Research on jailbreak has highlighted the safety issues of LLMs. However, prior studies have predominantly focused on single-turn dialogue, ignoring the potential complexities and risks presented by multi-turn dialogue, a crucial mode through which humans derive information from LLMs. In this paper, we argue that humans could exploit multi-turn dialogue to induce LLMs into generating harmful information. LLMs may not intend to reject cautionary or borderline unsafe queries, even if each turn is closely served for one malicious purpose in a multi-turn dialogue. Therefore, by decomposing an unsafe query into several sub-queries for multi-turn dialogue, we induced LLMs to answer harmful sub-questions incrementally, culminating in an overall harmful response. Our experiments, conducted across a wide range of LLMs, indicate current inadequacies in the safety mechanisms of LLMs in multi-turn dialogue. Our findings expose vulnerabilities of LLMs in complex scenarios involving multi-turn dialogue, presenting new challenges for the safety of LLMs.

Computation and Language

Mini-Ensemble Low-Rank Adapters for Parameter-Efficient Fine-Tuning

Parameter-efficient fine-tuning (PEFT) is a popular method for tailoring pre-trained large language models (LLMs), especially as the models' scale and the diversity of tasks increase. Low-rank adaptation (LoRA) is based on the idea that the adaptation process is intrinsically low-dimensional, i.e., significant model changes can be represented with relatively few parameters. However, decreasing the rank encounters challenges with generalization errors for specific tasks when compared to full-parameter fine-tuning. We present MELoRA, a mini-ensemble low-rank adapters that uses fewer trainable parameters while maintaining a higher rank, thereby offering improved performance potential. The core idea is to freeze original pretrained weights and train a group of mini LoRAs with only a small number of parameters. This can capture a significant degree of diversity among mini LoRAs, thus promoting better generalization ability. We conduct a theoretical analysis and empirical studies on various NLP tasks. Our experimental results show that, compared to LoRA, MELoRA achieves better performance with 8 times fewer trainable parameters on natural language understanding tasks and 36 times fewer trainable parameters on instruction following tasks, which demonstrates the effectiveness of MELoRA.

Computation and Language

Can LLM Generate Culturally Relevant Commonsense QA Data? Case Study in Indonesian and Sundanese

Large Language Models (LLMs) are increasingly being used to generate synthetic data for training and evaluating models. However, it is unclear whether they can generate a good quality of question answering (QA) dataset that incorporates knowledge and cultural nuance embedded in a language, especially for low-resource languages. In this study, we investigate the effectiveness of using LLMs in generating culturally relevant commonsense QA datasets for Indonesian and Sundanese languages. To do so, we create datasets for these languages using various methods involving both LLMs and human annotators. Our experiments show that the current best-performing LLM, GPT-4 Turbo, is capable of generating questions with adequate knowledge in Indonesian but not in Sundanese, highlighting the performance discrepancy between medium- and lower-resource languages. We also benchmark various LLMs on our generated datasets and find that they perform better on the LLM-generated datasets compared to those created by humans.

Computation and Language

Probing Multimodal Large Language Models for Global and Local Semantic Representation

The success of large language models has inspired researchers to transfer their exceptional representing ability to other modalities. Several recent works leverage image-caption alignment datasets to train multimodal large language models (MLLMs), which achieve state-of-the-art performance on image-to-text tasks. However, there are very few studies exploring whether MLLMs truly understand the complete image information, i.e., global information, or if they can only capture some local object information. In this study, we find that the intermediate layers of models can encode more global semantic information, whose representation vectors perform better on visual-language entailment tasks, rather than the topmost layers. We further probe models for local semantic representation through object detection tasks. And we draw a conclusion that the topmost layers may excessively focus on local information, leading to a diminished ability to encode global information.

Computation and Language

SKT5SciSumm -- A Hybrid Generative Approach for Multi-Document Scientific Summarization

Summarization for scientific text has shown significant benefits both for the research community and human society. Given the fact that the nature of scientific text is distinctive and the input of the multi-document summarization task is substantially long, the task requires sufficient embedding generation and text truncation without losing important information. To tackle these issues, in this paper, we propose SKT5SciSumm - a hybrid framework for multi-document scientific summarization (MDSS). We leverage the Sentence-Transformer version of Scientific Paper Embeddings using Citation-Informed Transformers (SPECTER) to encode and represent textual sentences, allowing for efficient extractive summarization using k-means clustering. We employ the T5 family of models to generate abstractive summaries using extracted sentences. SKT5SciSumm achieves state-of-the-art performance on the Multi-XScience dataset. Through extensive experiments and evaluation, we showcase the benefits of our model by using less complicated models to achieve remarkable results, thereby highlighting its potential in advancing the field of multi-document summarization for scientific text.

Computation and Language

Unsupervised multiple choices question answering via universal corpus

Unsupervised question answering is a promising yet challenging task, which alleviates the burden of building large-scale annotated data in a new domain. It motivates us to study the unsupervised multiple-choice question answering (MCQA) problem. In this paper, we propose a novel framework designed to generate synthetic MCQA data barely based on contexts from the universal domain without relying on any form of manual annotation. Possible answers are extracted and used to produce related questions, then we leverage both named entities (NE) and knowledge graphs to discover plausible distractors to form complete synthetic samples. Experiments on multiple MCQA datasets demonstrate the effectiveness of our method.

Computation and Language

RECOST: External Knowledge Guided Data-efficient Instruction Tuning

In the current landscape of large language models (LLMs), the process of instruction tuning serves as an essential step. Considering the high computing power overhead, data-efficient instruction tuning was proposed to reduce the training data size in this process, aiming at selecting high-quality instructional data. Nevertheless, we argue that most current data-efficient instruction-tuning methods are highly dependent on the quality of the original instruction-tuning dataset. When it comes to datasets synthesized by LLMs, a common scenario in this field, dirty samples will even be selected with a higher probability than other samples. To address these challenges, we utilized external knowledge (relevant examples or paragraphs) to evaluate those samples synthesized by LLMs with an in-context-based relative predictive entropy. Based on the new metric, we proposed a framework, dubbed as \textbf{RECOST}, which integrates external-knowledge-base re-ranking and diversity-consistent sampling into a single pipeline. Through extensive experiments on several synthetic datasets (Alpaca and Alpaca-gpt4), we demonstrate the effectiveness of our method and achieve even better results with only \textbf{1\%} of the full dataset.

Computation and Language

SoFA: Shielded On-the-fly Alignment via Priority Rule Following

The alignment problem in Large Language Models (LLMs) involves adapting them to the broad spectrum of human values. This requirement challenges existing alignment methods due to diversity of preferences and regulatory standards. This paper introduces a novel alignment paradigm, priority rule following, which defines rules as the primary control mechanism in each dialog, prioritizing them over user instructions. Our preliminary analysis reveals that even the advanced LLMs, such as GPT-4, exhibit shortcomings in understanding and prioritizing the rules. Therefore, we present PriorityDistill, a semi-automated approach for distilling priority following signals from LLM simulations to ensure robust rule integration and adherence. Our experiments show that this method not only effectively minimizes misalignments utilizing only one general rule but also adapts smoothly to various unseen rules, ensuring they are shielded from hijacking and that the model responds appropriately.

Computation and Language

A Dataset for Metaphor Detection in Early Medieval Hebrew Poetry

There is a large volume of late antique and medieval Hebrew texts. They represent a crucial linguistic and cultural bridge between Biblical and modern Hebrew. Poetry is prominent in these texts and one of its main haracteristics is the frequent use of metaphor. Distinguishing figurative and literal language use is a major task for scholars of the Humanities, especially in the fields of literature, linguistics, and hermeneutics. This paper presents a new, challenging dataset of late antique and medieval Hebrew poetry with expert annotations of metaphor, as well as some baseline results, which we hope will facilitate further research in this area.

Computation and Language

KoDialogBench: Evaluating Conversational Understanding of Language Models with Korean Dialogue Benchmark

As language models are often deployed as chatbot assistants, it becomes a virtue for models to engage in conversations in a user's first language. While these models are trained on a wide range of languages, a comprehensive evaluation of their proficiency in low-resource languages such as Korean has been lacking. In this work, we introduce KoDialogBench, a benchmark designed to assess language models' conversational capabilities in Korean. To this end, we collect native Korean dialogues on daily topics from public sources, or translate dialogues from other languages. We then structure these conversations into diverse test datasets, spanning from dialogue comprehension to response selection tasks. Leveraging the proposed benchmark, we conduct extensive evaluations and analyses of various language models to measure a foundational understanding of Korean dialogues. Experimental results indicate that there exists significant room for improvement in models' conversation skills. Furthermore, our in-depth comparisons across different language models highlight the effectiveness of recent training techniques in enhancing conversational proficiency. We anticipate that KoDialogBench will promote the progress towards conversation-aware Korean language models.

Computation and Language

FairBelief -- Assessing Harmful Beliefs in Language Models

Language Models (LMs) have been shown to inherit undesired biases that might hurt minorities and underrepresented groups if such systems were integrated into real-world applications without careful fairness auditing. This paper proposes FairBelief, an analytical approach to capture and assess beliefs, i.e., propositions that an LM may embed with different degrees of confidence and that covertly influence its predictions. With FairBelief, we leverage prompting to study the behavior of several state-of-the-art LMs across different previously neglected axes, such as model scale and likelihood, assessing predictions on a fairness dataset specifically designed to quantify LMs' outputs' hurtfulness. Finally, we conclude with an in-depth qualitative assessment of the beliefs emitted by the models. We apply FairBelief to English LMs, revealing that, although these architectures enable high performances on diverse natural language processing tasks, they show hurtful beliefs about specific genders. Interestingly, training procedure and dataset, model scale, and architecture induce beliefs of different degrees of hurtfulness.

Computation and Language

Spot the bot: Coarse-Grained Partition of Semantic Paths for Bots and Humans

Nowadays, technology is rapidly advancing: bots are writing comments, articles, and reviews. Due to this fact, it is crucial to know if the text was written by a human or by a bot. This paper focuses on comparing structures of the coarse-grained partitions of semantic paths for human-written and bot-generated texts. We compare the clusterizations of datasets of n-grams from literary texts and texts generated by several bots. The hypothesis is that the structures and clusterizations are different. Our research supports the hypothesis. As the semantic structure may be different for different languages, we investigate Russian, English, German, and Vietnamese languages.

Computation and Language

Benchmarking GPT-4 on Algorithmic Problems: A Systematic Evaluation of Prompting Strategies

Large Language Models (LLMs) have revolutionized the field of Natural Language Processing thanks to their ability to reuse knowledge acquired on massive text corpora on a wide variety of downstream tasks, with minimal (if any) tuning steps. At the same time, it has been repeatedly shown that LLMs lack systematic generalization, which allows to extrapolate the learned statistical regularities outside the training distribution. In this work, we offer a systematic benchmarking of GPT-4, one of the most advanced LLMs available, on three algorithmic tasks characterized by the possibility to control the problem difficulty with two parameters. We compare the performance of GPT-4 with that of its predecessor (GPT-3.5) and with a variant of the Transformer-Encoder architecture recently introduced to solve similar tasks, the Neural Data Router. We find that the deployment of advanced prompting techniques allows GPT-4 to reach superior accuracy on all tasks, demonstrating that state-of-the-art LLMs constitute a very strong baseline also in challenging tasks that require systematic generalization.

Computation and Language

Investigating Continual Pretraining in Large Language Models: Insights and Implications

This paper studies the evolving domain of Continual Learning (CL) in large language models (LLMs), with a focus on developing strategies for efficient and sustainable training. Our primary emphasis is on continual domain-adaptive pretraining, a process designed to equip LLMs with the ability to integrate new information from various domains while retaining previously learned knowledge and enhancing cross-domain knowledge transfer without relying on domain-specific identification. Unlike previous studies, which mostly concentrate on a limited selection of tasks or domains and primarily aim to address the issue of forgetting, our research evaluates the adaptability and capabilities of LLMs to changing data landscapes in practical scenarios. To this end, we introduce a new benchmark designed to measure the adaptability of LLMs to these evolving data environments, offering a comprehensive framework for evaluation. We examine the impact of model size on learning efficacy and forgetting, as well as how the progression and similarity of emerging domains affect the knowledge transfer within these models. Our findings uncover several key insights: (i) when the sequence of domains shows semantic similarity, continual pretraining enables LLMs to better specialize in the current domain compared to stand-alone fine-tuning, (ii) training across a diverse range of domains enhances both backward and forward knowledge transfer, and (iii) smaller models are particularly sensitive to continual pretraining, showing the most significant rates of both forgetting and learning. We posit that our research marks a shift towards establishing a more realistic benchmark for investigating CL in LLMs, and has the potential to play a key role in guiding the direction of future research in the field.

Computation and Language

Consistency Matters: Explore LLMs Consistency From a Black-Box Perspective

Nowadays both commercial and open-source academic LLM have become the mainstream models of NLP. However, there is still a lack of research on LLM consistency, meaning that throughout the various stages of LLM research and deployment, its internal parameters and capabilities should remain unchanged. This issue exists in both the industrial and academic sectors. The solution to this problem is often time-consuming and labor-intensive, and there is also an additional cost of secondary deployment, resulting in economic and time losses. To fill this gap, we build an LLM consistency task dataset and design several baselines. Additionally, we choose models of diverse scales for the main experiments. Specifically, in the LightGBM experiment, we used traditional NLG metrics (i.e., ROUGE, BLEU, METEOR) as the features needed for model training. The final result exceeds the manual evaluation and GPT3.5 as well as other models in the main experiment, achieving the best performance. In the end, we use the best performing LightGBM model as the base model to build the evaluation tool, which can effectively assist in the deployment of business models. Our code and tool demo are available at https://github.com/heavenhellchen/Consistency.git

Computation and Language

Enhancing EEG-to-Text Decoding through Transferable Representations from Pre-trained Contrastive EEG-Text Masked Autoencoder

Reconstructing natural language from non-invasive electroencephalography (EEG) holds great promise as a language decoding technology for brain-computer interfaces (BCIs). However, EEG-based language decoding is still in its nascent stages, facing several technical issues such as: 1) Absence of a hybrid strategy that can effectively integrate cross-modality (between EEG and text) self-learning with intra-modality self-reconstruction of EEG features or textual sequences; 2) Under-utilization of large language models (LLMs) to enhance EEG-based language decoding. To address above issues, we propose the Contrastive EEG-Text Masked Autoencoder (CET-MAE), a novel model that orchestrates compound self-supervised learning across and within EEG and text through a dedicated multi-stream encoder. Furthermore, we develop a framework called E2T-PTR (EEG-to-Text decoding using Pretrained Transferable Representations), which leverages pre-trained modules alongside the EEG stream from CET-MAE and further enables an LLM (specifically BART) to decode text from EEG sequences. Comprehensive experiments conducted on the popular text-evoked EEG database, ZuCo, demonstrate the superiority of E2T-PTR, which outperforms the state-of-the-art in ROUGE-1 F1 and BLEU-4 scores by 8.34% and 32.21%, respectively. These results indicate significant advancements in the field and underscores the proposed framework's potential to enable more powerful and widespread BCI applications.

Computation and Language

Exploiting Emotion-Semantic Correlations for Empathetic Response Generation

Empathetic response generation aims to generate empathetic responses by understanding the speaker's emotional feelings from the language of dialogue. Recent methods capture emotional words in the language of communicators and construct them as static vectors to perceive nuanced emotions. However, linguistic research has shown that emotional words in language are dynamic and have correlations with other grammar semantic roles, i.e., words with semantic meanings, in grammar. Previous methods overlook these two characteristics, which easily lead to misunderstandings of emotions and neglect of key semantics. To address this issue, we propose a dynamical Emotion-Semantic Correlation Model (ESCM) for empathetic dialogue generation tasks. ESCM constructs dynamic emotion-semantic vectors through the interaction of context and emotions. We introduce dependency trees to reflect the correlations between emotions and semantics. Based on dynamic emotion-semantic vectors and dependency trees, we propose a dynamic correlation graph convolutional network to guide the model in learning context meanings in dialogue and generating empathetic responses. Experimental results on the EMPATHETIC-DIALOGUES dataset show that ESCM understands semantics and emotions more accurately and expresses fluent and informative empathetic responses. Our analysis results also indicate that the correlations between emotions and semantics are frequently used in dialogues, which is of great significance for empathetic perception and expression.

Computation and Language

Deep Learning Based Named Entity Recognition Models for Recipes

Food touches our lives through various endeavors, including flavor, nourishment, health, and sustainability. Recipes are cultural capsules transmitted across generations via unstructured text. Automated protocols for recognizing named entities, the building blocks of recipe text, are of immense value for various applications ranging from information extraction to novel recipe generation. Named entity recognition is a technique for extracting information from unstructured or semi-structured data with known labels. Starting with manually-annotated data of 6,611 ingredient phrases, we created an augmented dataset of 26,445 phrases cumulatively. Simultaneously, we systematically cleaned and analyzed ingredient phrases from RecipeDB, the gold-standard recipe data repository, and annotated them using the Stanford NER. Based on the analysis, we sampled a subset of 88,526 phrases using a clustering-based approach while preserving the diversity to create the machine-annotated dataset. A thorough investigation of NER approaches on these three datasets involving statistical, fine-tuning of deep learning-based language models and few-shot prompting on large language models (LLMs) provides deep insights. We conclude that few-shot prompting on LLMs has abysmal performance, whereas the fine-tuned spaCy-transformer emerges as the best model with macro-F1 scores of 95.9%, 96.04%, and 95.71% for the manually-annotated, augmented, and machine-annotated datasets, respectively.

Computation and Language

Training-Free Long-Context Scaling of Large Language Models

The ability of Large Language Models (LLMs) to process and generate coherent text is markedly weakened when the number of input tokens exceeds their pretraining length. Given the expensive overhead of finetuning large-scale models with longer sequences, we propose Dual Chunk Attention (DCA), which enables Llama2 70B to support context windows of more than 100k tokens without continual training. By decomposing the attention computation for long sequences into chunk-based modules, DCA manages to effectively capture the relative positional information of tokens within the same chunk (Intra-Chunk) and across distinct chunks (Inter-Chunk), as well as integrates seamlessly with Flash Attention. In addition to its impressive extrapolation capability, DCA achieves performance on practical long-context tasks that is comparable to or even better than that of finetuned models. When compared with proprietary models, our training-free 70B model attains 94% of the performance of gpt-3.5-16k, indicating it is a viable open-source alternative. All code and data used in this work are released at \url{https://github.com/HKUNLP/ChunkLlama}.

Computation and Language

Can GPT-4 Identify Propaganda? Annotation and Detection of Propaganda Spans in News Articles

The use of propaganda has spiked on mainstream and social media, aiming to manipulate or mislead users. While efforts to automatically detect propaganda techniques in textual, visual, or multimodal content have increased, most of them primarily focus on English content. The majority of the recent initiatives targeting medium to low-resource languages produced relatively small annotated datasets, with a skewed distribution, posing challenges for the development of sophisticated propaganda detection models. To address this challenge, we carefully develop the largest propaganda dataset to date, ArPro, comprised of 8K paragraphs from newspaper articles, labeled at the text span level following a taxonomy of 23 propagandistic techniques. Furthermore, our work offers the first attempt to understand the performance of large language models (LLMs), using GPT-4, for fine-grained propaganda detection from text. Results showed that GPT-4's performance degrades as the task moves from simply classifying a paragraph as propagandistic or not, to the fine-grained task of detecting propaganda techniques and their manifestation in text. Compared to models fine-tuned on the dataset for propaganda detection at different classification granularities, GPT-4 is still far behind. Finally, we evaluate GPT-4 on a dataset consisting of six other languages for span detection, and results suggest that the model struggles with the task across languages. Our dataset and resources will be released to the community.

Computation and Language

Prescribing Large Language Models for Perioperative Care: What's The Right Dose for Pre-trained Models?

Postoperative risk predictions can inform effective perioperative care management and planning. We aimed to assess whether clinical large language models (LLMs) can predict postoperative risks using clinical texts with various training strategies. The main cohort involved 84,875 records from Barnes Jewish Hospital (BJH) system between 2018 and 2021. Methods were replicated on Beth Israel Deaconess's MIMIC dataset. Both studies had mean duration of follow-up based on the length of postoperative ICU stay less than 7 days. For the BJH dataset, outcomes included 30-day mortality, pulmonary embolism (PE) and pneumonia. Three domain adaptation and finetuning strategies were implemented for BioGPT, ClinicalBERT and BioClinicalBERT: self-supervised objectives; incorporating labels with semi-supervised fine-tuning; and foundational modelling through multi-task learning. Model performance was compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision recall curve (AUPRC) for classification tasks, and mean squared error (MSE) and R2 for regression tasks. Pre-trained LLMs outperformed traditional word embeddings, with absolute maximal gains of 38.3% for AUROC and 14% for AUPRC. Adapting models further improved performance: (1) self-supervised finetuning by 3.2% for AUROC and 1.5% for AUPRC; (2) semi-supervised finetuning by 1.8% for AUROC and 2% for AUPRC, compared to self-supervised finetuning; (3) foundational modelling by 3.6% for AUROC and 2.6% for AUPRC, compared to self-supervised finetuning. Pre-trained clinical LLMs offer opportunities for postoperative risk predictions in unforeseen data, with peaks in foundational models indicating the potential of task-agnostic learning towards the generalizability of LLMs in perioperative care.

Computation and Language

REAR: A Relevance-Aware Retrieval-Augmented Framework for Open-Domain Question Answering

Considering the limited internal parametric knowledge, retrieval-augmented generation (RAG) has been widely used to extend the knowledge scope of large language models (LLMs). Despite the extensive efforts on RAG research, in existing methods, LLMs cannot precisely assess the relevance of retrieved documents, thus likely leading to misleading or even incorrect utilization of external knowledge (i.e., retrieved documents). To address this issue, in this paper, we propose REAR, a RElevance-Aware Retrieval-augmented approach for open-domain question answering (QA). As the key motivation, we aim to enhance the self-awareness of source relevance for LLMs, so as to adaptively utilize external knowledge in RAG systems. Specially, we develop a new architecture for LLM based RAG system, by incorporating a specially designed rank head that precisely assesses the relevance of retrieved documents. Furthermore, we propose an improved training method based on bi-granularity relevance fusion and noise-resistant training. By combining the improvements in both architecture and training, our proposed REAR can better utilize external knowledge by effectively perceiving the relevance of retrieved documents. Experiments on four open-domain QA tasks show that REAR significantly outperforms previous a number of competitive RAG approaches. Our code and data can be accessed at https://github.com/RUCAIBox/REAR.

Computation and Language

Extreme Miscalibration and the Illusion of Adversarial Robustness

Deep learning-based Natural Language Processing (NLP) models are vulnerable to adversarial attacks, where small perturbations can cause a model to misclassify. Adversarial Training (AT) is often used to increase model robustness. However, we have discovered an intriguing phenomenon: deliberately or accidentally miscalibrating models masks gradients in a way that interferes with adversarial attack search methods, giving rise to an apparent increase in robustness. We show that this observed gain in robustness is an illusion of robustness (IOR), and demonstrate how an adversary can perform various forms of test-time temperature calibration to nullify the aforementioned interference and allow the adversarial attack to find adversarial examples. Hence, we urge the NLP community to incorporate test-time temperature scaling into their robustness evaluations to ensure that any observed gains are genuine. Finally, we show how the temperature can be scaled during \textit{training} to improve genuine robustness.

Computation and Language

Latent Attention for Linear Time Transformers

The time complexity of the standard attention mechanism in a transformer scales quadratically with the length of the sequence. We introduce a method to reduce this to linear scaling with time, based on defining attention via latent vectors. The method is readily usable as a drop-in replacement for the standard attention mechanism. Our "Latte Transformer" model can be implemented for both bidirectional and unidirectional tasks, with the causal version allowing a recurrent implementation which is memory and time-efficient during inference of language generation tasks. Whilst next token prediction scales linearly with the sequence length for a standard transformer, a Latte Transformer requires constant time to compute the next token. The empirical performance of our method is comparable to standard attention, yet allows scaling to context windows much larger than practical in standard attention.

Computation and Language

Predict the Next Word: <Humans exhibit uncertainty in this task and language models _____>

Language models (LMs) are statistical models trained to assign probability to human-generated text. As such, it is reasonable to question whether they approximate linguistic variability exhibited by humans well. This form of statistical assessment is difficult to perform at the passage level, for it requires acceptability judgements (i.e., human evaluation) or a robust automated proxy (which is non-trivial). At the word level, however, given some context, samples from an LM can be assessed via exact matching against a prerecorded dataset of alternative single-word continuations of the available context. We exploit this fact and evaluate the LM's ability to reproduce variability that humans (in particular, a population of English speakers) exhibit in the 'next word prediction' task. This can be seen as assessing a form of calibration, which, in the context of text classification, Baan et al. (2022) termed calibration to human uncertainty. We assess GPT2, BLOOM and ChatGPT and find that they exhibit fairly low calibration to human uncertainty. We also verify the failure of expected calibration error (ECE) to reflect this, and as such, advise the community against relying on it in this setting.

Computation and Language

Retrieval is Accurate Generation

Standard language models generate text by selecting tokens from a fixed, finite, and standalone vocabulary. We introduce a novel method that selects context-aware phrases from a collection of supporting documents. One of the most significant challenges for this paradigm shift is determining the training oracles, because a string of text can be segmented in various ways and each segment can be retrieved from numerous possible documents. To address this, we propose to initialize the training oracles using linguistic heuristics and, more importantly, bootstrap the oracles through iterative self-reinforcement. Extensive experiments show that our model not only outperforms standard language models on a variety of knowledge-intensive tasks but also demonstrates improved generation quality in open-ended text generation. For instance, compared to the standard language model counterpart, our model raises the accuracy from 23.47% to 36.27% on OpenbookQA, and improves the MAUVE score from 42.61% to 81.58% in open-ended text generation. Remarkably, our model also achieves the best performance and the lowest latency among several retrieval-augmented baselines. In conclusion, we assert that retrieval is more accurate generation and hope that our work will encourage further research on this new paradigm shift.

Computation and Language

Unleashing the Potential of Large Language Models as Prompt Optimizers: An Analogical Analysis with Gradient-based Model Optimizers

Automatic prompt optimization is an important approach to improving the performance of large language models (LLMs). Recent research demonstrates the potential of using LLMs as prompt optimizers, which can generate improved task prompts via iterative refinement. In this paper, we propose a novel perspective to investigate the design of LLM-based prompt optimizers, by drawing an analogy with gradient-based model optimizers. To connect these two approaches, we identify two pivotal factors in model parameter learning: update direction and update method. Focused on the two aspects, we borrow the theoretical framework and learning methods from gradient-based optimization to design improved strategies for LLM-based prompt optimizers. By systematically analyzing a rich set of improvement strategies, we further develop a capable Gradient-inspired LLM-based Prompt Optimizer called GPO. At each step, it first retrieves relevant prompts from the optimization trajectory as the update direction. Then, it utilizes the generation-based refinement strategy to perform the update, while controlling the edit distance through a cosine-based decay strategy. Extensive experiments demonstrate the effectiveness and efficiency of GPO. In particular, GPO brings an additional improvement of up to 56.8% on Big-Bench Hard and 55.3% on MMLU compared to baseline methods.

Computation and Language

Linguistic Knowledge Can Enhance Encoder-Decoder Models (If You Let It)

In this paper, we explore the impact of augmenting pre-trained Encoder-Decoder models, specifically T5, with linguistic knowledge for the prediction of a target task. In particular, we investigate whether fine-tuning a T5 model on an intermediate task that predicts structural linguistic properties of sentences modifies its performance in the target task of predicting sentence-level complexity. Our study encompasses diverse experiments conducted on Italian and English datasets, employing both monolingual and multilingual T5 models at various sizes. Results obtained for both languages and in cross-lingual configurations show that linguistically motivated intermediate fine-tuning has generally a positive impact on target task performance, especially when applied to smaller models and in scenarios with limited data availability.

Computation and Language

Neural Automated Writing Evaluation with Corrective Feedback

The utilization of technology in second language learning and teaching has become ubiquitous. For the assessment of writing specifically, automated writing evaluation (AWE) and grammatical error correction (GEC) have become immensely popular and effective methods for enhancing writing proficiency and delivering instant and individualized feedback to learners. By leveraging the power of natural language processing (NLP) and machine learning algorithms, AWE and GEC systems have been developed separately to provide language learners with automated corrective feedback and more accurate and unbiased scoring that would otherwise be subject to examiners. In this paper, we propose an integrated system for automated writing evaluation with corrective feedback as a means of bridging the gap between AWE and GEC results for second language learners. This system enables language learners to simulate the essay writing tests: a student writes and submits an essay, and the system returns the assessment of the writing along with suggested grammatical error corrections. Given that automated scoring and grammatical correction are more efficient and cost-effective than human grading, this integrated system would also alleviate the burden of manually correcting innumerable essays.

Computation and Language

Fine-Grained Natural Language Inference Based Faithfulness Evaluation for Diverse Summarisation Tasks

We study existing approaches to leverage off-the-shelf Natural Language Inference (NLI) models for the evaluation of summary faithfulness and argue that these are sub-optimal due to the granularity level considered for premises and hypotheses. That is, the smaller content unit considered as hypothesis is a sentence and premises are made up of a fixed number of document sentences. We propose a novel approach, namely InFusE, that uses a variable premise size and simplifies summary sentences into shorter hypotheses. Departing from previous studies which focus on single short document summarisation, we analyse NLI based faithfulness evaluation for diverse summarisation tasks. We introduce DiverSumm, a new benchmark comprising long form summarisation (long documents and summaries) and diverse summarisation tasks (e.g., meeting and multi-document summarisation). In experiments, InFusE obtains superior performance across the different summarisation tasks. Our code and data are available at https://github.com/HJZnlp/infuse.

Computation and Language

From Text Segmentation to Smart Chaptering: A Novel Benchmark for Structuring Video Transcriptions

Text segmentation is a fundamental task in natural language processing, where documents are split into contiguous sections. However, prior research in this area has been constrained by limited datasets, which are either small in scale, synthesized, or only contain well-structured documents. In this paper, we address these limitations by introducing a novel benchmark YTSeg focusing on spoken content that is inherently more unstructured and both topically and structurally diverse. As part of this work, we introduce an efficient hierarchical segmentation model MiniSeg, that outperforms state-of-the-art baselines. Lastly, we expand the notion of text segmentation to a more practical "smart chaptering" task that involves the segmentation of unstructured content, the generation of meaningful segment titles, and a potential real-time application of the models.

Computation and Language

Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Data

Quantitative reasoning is a critical skill to analyze data, yet the assessment of such ability remains limited. To address this gap, we introduce the Quantitative Reasoning with Data (QRData) benchmark, aiming to evaluate Large Language Models' capability in statistical and causal reasoning with real-world data. The benchmark comprises a carefully constructed dataset of 411 questions accompanied by data sheets from textbooks, online learning materials, and academic papers. To compare models' quantitative reasoning abilities on data and text, we enrich the benchmark with an auxiliary set of 290 text-only questions, namely QRText. We evaluate natural language reasoning, program-based reasoning, and agent reasoning methods including Chain-of-Thought, Program-of-Thoughts, ReAct, and code interpreter assistants on diverse models. The strongest model GPT-4 achieves an accuracy of 58%, which has a large room for improvement. Among open-source models, Deepseek-coder-instruct, a code LLM pretrained on 2T tokens, gets the highest accuracy of 37%. Analysis reveals that models encounter difficulties in data analysis and causal reasoning, and struggle in using causal knowledge and provided data simultaneously. Code and data are in https://github.com/xxxiaol/QRData.

Computation and Language

Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMs

Due to the widespread use of large language models (LLMs) in ubiquitous systems, we need to understand whether they embed a specific worldview and what these views reflect. Recent studies report that, prompted with political questionnaires, LLMs show left-liberal leanings. However, it is as yet unclear whether these leanings are reliable (robust to prompt variations) and whether the leaning is consistent across policies and political leaning. We propose a series of tests which assess the reliability and consistency of LLMs' stances on political statements based on a dataset of voting-advice questionnaires collected from seven EU countries and annotated for policy domains. We study LLMs ranging in size from 7B to 70B parameters and find that their reliability increases with parameter count. Larger models show overall stronger alignment with left-leaning parties but differ among policy programs: They evince a (left-wing) positive stance towards environment protection, social welfare but also (right-wing) law and order, with no consistent preferences in foreign policy, migration, and economy.

Computation and Language

NextLevelBERT: Investigating Masked Language Modeling with Higher-Level Representations for Long Documents

While (large) language models have significantly improved over the last years, they still struggle to sensibly process long sequences found, e.g., in books, due to the quadratic scaling of the underlying attention mechanism. To address this, we propose NextLevelBERT, a Masked Language Model operating not on tokens, but on higher-level semantic representations in the form of text embeddings. We pretrain NextLevelBERT to predict the vector representation of entire masked text chunks and evaluate the effectiveness of the resulting document vectors on three task types: 1) Semantic Textual Similarity via zero-shot document embeddings, 2) Long document classification, 3) Multiple-choice question answering. We find that next level Masked Language Modeling is an effective technique to tackle long-document use cases and can outperform much larger embedding models as long as the required level of detail is not too high. We make model and code available.

Computation and Language

RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representations

Individual neurons participate in the representation of multiple high-level concepts. To what extent can different interpretability methods successfully disentangle these roles? To help address this question, we introduce RAVEL (Resolving Attribute-Value Entanglements in Language Models), a dataset that enables tightly controlled, quantitative comparisons between a variety of existing interpretability methods. We use the resulting conceptual framework to define the new method of Multi-task Distributed Alignment Search (MDAS), which allows us to find distributed representations satisfying multiple causal criteria. With Llama2-7B as the target language model, MDAS achieves state-of-the-art results on RAVEL, demonstrating the importance of going beyond neuron-level analyses to identify features distributed across activations. We release our benchmark at https://github.com/explanare/ravel.

Computation and Language

AmbigNLG: Addressing Task Ambiguity in Instruction for NLG

In this study, we introduce AmbigNLG, a new task designed to tackle the challenge of task ambiguity in instructions for Natural Language Generation (NLG) tasks. Despite the impressive capabilities of Large Language Models (LLMs) in understanding and executing a wide range of tasks through natural language interaction, their performance is significantly hindered by the ambiguity present in real-world instructions. To address this, AmbigNLG seeks to identify and mitigate such ambiguities, aiming to refine instructions to match user expectations better. We introduce a dataset, AmbigSNI-NLG, consisting of 2,500 instances, and develop an ambiguity taxonomy for categorizing and annotating instruction ambiguities. Our approach demonstrates substantial improvements in text generation quality, highlighting the critical role of clear and specific instructions in enhancing LLM performance in NLG tasks.

Computation and Language

Tower: An Open Multilingual Large Language Model for Translation-Related Tasks

While general-purpose large language models (LLMs) demonstrate proficiency on multiple tasks within the domain of translation, approaches based on open LLMs are competitive only when specializing on a single task. In this paper, we propose a recipe for tailoring LLMs to multiple tasks present in translation workflows. We perform continued pretraining on a multilingual mixture of monolingual and parallel data, creating TowerBase, followed by finetuning on instructions relevant for translation processes, creating TowerInstruct. Our final model surpasses open alternatives on several tasks relevant to translation workflows and is competitive with general-purpose closed LLMs. To facilitate future research, we release the Tower models, our specialization dataset, an evaluation framework for LLMs focusing on the translation ecosystem, and a collection of model generations, including ours, on our benchmark.

Computation and Language

Evaluating Very Long-Term Conversational Memory of LLM Agents

Existing works on long-term open-domain dialogues focus on evaluating model responses within contexts spanning no more than five chat sessions. Despite advancements in long-context large language models (LLMs) and retrieval augmented generation (RAG) techniques, their efficacy in very long-term dialogues remains unexplored. To address this research gap, we introduce a machine-human pipeline to generate high-quality, very long-term dialogues by leveraging LLM-based agent architectures and grounding their dialogues on personas and temporal event graphs. Moreover, we equip each agent with the capability of sharing and reacting to images. The generated conversations are verified and edited by human annotators for long-range consistency and grounding to the event graphs. Using this pipeline, we collect LoCoMo, a dataset of very long-term conversations, each encompassing 300 turns and 9K tokens on avg., over up to 35 sessions. Based on LoCoMo, we present a comprehensive evaluation benchmark to measure long-term memory in models, encompassing question answering, event summarization, and multi-modal dialogue generation tasks. Our experimental results indicate that LLMs exhibit challenges in understanding lengthy conversations and comprehending long-range temporal and causal dynamics within dialogues. Employing strategies like long-context LLMs or RAG can offer improvements but these models still substantially lag behind human performance.

Computation and Language

Towards Optimal Learning of Language Models

This work studies the general principles of improving the learning of language models (LMs), which aims at reducing the necessary training steps for achieving superior performance. Specifically, we present a theory for the optimal learning of LMs. We first propose an objective that optimizes LM learning by maximizing the data compression ratio in an "LM-training-as-lossless-compression" view. Then, we derive a theorem, named Learning Law, to reveal the properties of the dynamics in the optimal learning process under our objective. The theorem is then validated by experiments on a linear classification and a real-world language modeling task. Finally, we empirically verify that the optimal learning of LMs essentially stems from the improvement of the coefficients in the scaling law of LMs, indicating great promise and significance for designing practical learning acceleration methods. Our code can be found at https://aka.ms/LearningLaw.

Computation and Language

Massive Activations in Large Language Models

We observe an empirical phenomenon in Large Language Models (LLMs) -- very few activations exhibit significantly larger values than others (e.g., 100,000 times larger). We call them massive activations. First, we demonstrate the widespread existence of massive activations across various LLMs and characterize their locations. Second, we find their values largely stay constant regardless of the input, and they function as indispensable bias terms in LLMs. Third, these massive activations lead to the concentration of attention probabilities to their corresponding tokens, and further, implicit bias terms in the self-attention output. Last, we also study massive activations in Vision Transformers.

Computation and Language

The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits

Recent research, such as BitNet, is paving the way for a new era of 1-bit Large Language Models (LLMs). In this work, we introduce a 1-bit LLM variant, namely BitNet b1.58, in which every single parameter (or weight) of the LLM is ternary {-1, 0, 1}. It matches the full-precision (i.e., FP16 or BF16) Transformer LLM with the same model size and training tokens in terms of both perplexity and end-task performance, while being significantly more cost-effective in terms of latency, memory, throughput, and energy consumption. More profoundly, the 1.58-bit LLM defines a new scaling law and recipe for training new generations of LLMs that are both high-performance and cost-effective. Furthermore, it enables a new computation paradigm and opens the door for designing specific hardware optimized for 1-bit LLMs.

Computation and Language

TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space

Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks. However, they sometimes suffer from producing hallucinations, particularly in cases where they may generate untruthful responses despite possessing the correct knowledge. In this paper, we propose TruthX, an inference-time method to elicit the truthfulness of LLMs by editing their internal representations in truthful space. TruthX employs an auto-encoder to map LLM's representations into semantic and truthful latent spaces respectively, and applies contrastive learning to identify a truthful editing direction within the truthful space. During inference, by editing LLM's internal representations in truthful space, TruthX effectively enhances the truthfulness of LLMs. Experiments show that TruthX effectively improves the truthfulness of 13 advanced LLMs by an average of 20% on TruthfulQA benchmark. Further analyses suggest that the truthful space acquired by TruthX plays a pivotal role in controlling LLM to produce truthful or hallucinatory responses.

Computation and Language

Stable LM 2 1.6B Technical Report

We introduce StableLM 2 1.6B, the first in a new generation of our language model series. In this technical report, we present in detail the data and training procedure leading to the base and instruction-tuned versions of StableLM 2 1.6B. The weights for both models are available via Hugging Face for anyone to download and use. The report contains thorough evaluations of these models, including zero- and few-shot benchmarks, multilingual benchmarks, and the MT benchmark focusing on multi-turn dialogues. At the time of publishing this report, StableLM 2 1.6B was the state-of-the-art open model under 2B parameters by a significant margin. Given its appealing small size, we also provide throughput measurements on a number of edge devices. In addition, we open source several quantized checkpoints and provide their performance metrics compared to the original model.

Computation and Language

Follow My Instruction and Spill the Beans: Scalable Data Extraction from Retrieval-Augmented Generation Systems

Retrieval-Augmented Generation (RAG) improves pre-trained models by incorporating external knowledge at test time to enable customized adaptation. We study the risk of datastore leakage in Retrieval-In-Context RAG Language Models (LMs). We show that an adversary can exploit LMs' instruction-following capabilities to easily extract text data verbatim from the datastore of RAG systems built with instruction-tuned LMs via prompt injection. The vulnerability exists for a wide range of modern LMs that span Llama2, Mistral/Mixtral, Vicuna, SOLAR, WizardLM, Qwen1.5, and Platypus2, and the exploitability exacerbates as the model size scales up. Extending our study to production RAG models GPTs, we design an attack that can cause datastore leakage with a 100% success rate on 25 randomly selected customized GPTs with at most 2 queries, and we extract text data verbatim at a rate of 41% from a book of 77,000 words and 3% from a corpus of 1,569,000 words by prompting the GPTs with only 100 queries generated by themselves.

Computation and Language

BlendSQL: A Scalable Dialect for Unifying Hybrid Question Answering in Relational Algebra

Many existing end-to-end systems for hybrid question answering tasks can often be boiled down to a "prompt-and-pray" paradigm, where the user has limited control and insight into the intermediate reasoning steps used to achieve the final result. Additionally, due to the context size limitation of many transformer-based LLMs, it is often not reasonable to expect that the full structured and unstructured context will fit into a given prompt in a zero-shot setting, let alone a few-shot setting. We introduce BlendSQL, a superset of SQLite to act as a unified dialect for orchestrating reasoning across both unstructured and structured data. For hybrid question answering tasks involving multi-hop reasoning, we encode the full decomposed reasoning roadmap into a single interpretable BlendSQL query. Notably, we show that BlendSQL can scale to massive datasets and improve the performance of end-to-end systems while using 35% fewer tokens. Our code is available and installable as a package at https://github.com/parkervg/blendsql.

Computation and Language

JMLR: Joint Medical LLM and Retrieval Training for Enhancing Reasoning and Professional Question Answering Capability

With the explosive growth of medical data and the rapid development of artificial intelligence technology, precision medicine has emerged as a key to enhancing the quality and efficiency of healthcare services. In this context, Large Language Models (LLMs) play an increasingly vital role in medical knowledge acquisition and question-answering systems. To further improve the performance of these systems in the medical domain, we introduce an innovative method that jointly trains an Information Retrieval (IR) system and an LLM during the fine-tuning phase. This approach, which we call Joint Medical LLM and Retrieval Training (JMLR), is designed to overcome the challenges faced by traditional models in handling medical question-answering tasks. By employing a synchronized training mechanism, JMLR reduces the demand for computational resources and enhances the model's ability to leverage medical knowledge for reasoning and answering questions. Our experimental results demonstrate that JMLR-13B (81.2% on Amboos, 61.3% on MedQA) outperforms models using conventional pre-training and fine-tuning Meditron-70B (76.4% on AMBOSS, 60.3% on MedQA). For models of the same 7B scale, JMLR-7B(68.7% on Amboos, 51.7% on MedQA) significantly outperforms other public models (Meditron-7B: 50.1%, 47.9%), proving its superiority in terms of cost (our training time: 37 hours, traditional method: 144 hours), efficiency, and effectiveness in medical question-answering tasks. Through this work, we provide a new and efficient knowledge enhancement tool for healthcare, demonstrating the great potential of integrating IR and LLM training in precision medical information retrieval and question-answering systems.

Computation and Language

Researchy Questions: A Dataset of Multi-Perspective, Decompositional Questions for LLM Web Agents

Existing question answering (QA) datasets are no longer challenging to most powerful Large Language Models (LLMs). Traditional QA benchmarks like TriviaQA, NaturalQuestions, ELI5 and HotpotQA mainly study ``known unknowns'' with clear indications of both what information is missing, and how to find it to answer the question. Hence, good performance on these benchmarks provides a false sense of security. A yet unmet need of the NLP community is a bank of non-factoid, multi-perspective questions involving a great deal of unclear information needs, i.e. ``unknown uknowns''. We claim we can find such questions in search engine logs, which is surprising because most question-intent queries are indeed factoid. We present Researchy Questions, a dataset of search engine queries tediously filtered to be non-factoid, ``decompositional'' and multi-perspective. We show that users spend a lot of ``effort'' on these questions in terms of signals like clicks and session length, and that they are also challenging for GPT-4. We also show that ``slow thinking'' answering techniques, like decomposition into sub-questions shows benefit over answering directly. We release $\sim$ 100k Researchy Questions, along with the Clueweb22 URLs that were clicked.

Computation and Language

A Language Model based Framework for New Concept Placement in Ontologies

We investigate the task of inserting new concepts extracted from texts into an ontology using language models. We explore an approach with three steps: edge search which is to find a set of candidate locations to insert (i.e., subsumptions between concepts), edge formation and enrichment which leverages the ontological structure to produce and enhance the edge candidates, and edge selection which eventually locates the edge to be placed into. In all steps, we propose to leverage neural methods, where we apply embedding-based methods and contrastive learning with Pre-trained Language Models (PLMs) such as BERT for edge search, and adapt a BERT fine-tuning-based multi-label Edge-Cross-encoder, and Large Language Models (LLMs) such as GPT series, FLAN-T5, and Llama 2, for edge selection. We evaluate the methods on recent datasets created using the SNOMED CT ontology and the MedMentions entity linking benchmark. The best settings in our framework use fine-tuned PLM for search and a multi-label Cross-encoder for selection. Zero-shot prompting of LLMs is still not adequate for the task, and we propose explainable instruction tuning of LLMs for improved performance. Our study shows the advantages of PLMs and highlights the encouraging performance of LLMs that motivates future studies.

Computation and Language

Extracting Lexical Features from Dialects via Interpretable Dialect Classifiers

Identifying linguistic differences between dialects of a language often requires expert knowledge and meticulous human analysis. This is largely due to the complexity and nuance involved in studying various dialects. We present a novel approach to extract distinguishing lexical features of dialects by utilizing interpretable dialect classifiers, even in the absence of human experts. We explore both post-hoc and intrinsic approaches to interpretability, conduct experiments on Mandarin, Italian, and Low Saxon, and experimentally demonstrate that our method successfully identifies key language-specific lexical features that contribute to dialectal variations.

Computation and Language

LLM-Resistant Math Word Problem Generation via Adversarial Attacks

Large language models (LLMs) have significantly transformed the educational landscape. As current plagiarism detection tools struggle to keep pace with LLMs' rapid advancements, the educational community faces the challenge of assessing students' true problem-solving abilities in the presence of LLMs. In this work, we explore a new paradigm for ensuring fair evaluation -- generating adversarial examples which preserve the structure and difficulty of the original questions aimed for assessment, but are unsolvable by LLMs. Focusing on the domain of math word problems, we leverage abstract syntax trees to structurally generate adversarial examples that cause LLMs to produce incorrect answers by simply editing the numeric values in the problems. We conduct experiments on various open- and closed-source LLMs, quantitatively and qualitatively demonstrating that our method significantly degrades their math problem-solving ability. We identify shared vulnerabilities among LLMs and propose a cost-effective approach to attack high-cost models. Additionally, we conduct automatic analysis on math problems and investigate the cause of failure to guide future research on LLM's mathematical capability.

Computation and Language

Multitask Multilingual Model Adaptation with Featurized Low-Rank Mixtures

Adapting pretrained large language models (LLMs) to various downstream tasks in tens or hundreds of human languages is computationally expensive. Parameter-efficient fine-tuning (PEFT) significantly reduces the adaptation cost, by tuning only a small amount of parameters. However, directly applying PEFT methods such as LoRA (Hu et al., 2022) on diverse dataset mixtures could lead to suboptimal performance due to limited parameter capacity and negative interference among different datasets. In this work, we propose Featurized Low-rank Mixtures (FLix), a novel PEFT method designed for effective multitask multilingual tuning. FLix associates each unique dataset feature, such as the dataset's language or task, with its own low-rank weight update parameters. By composing feature-specific parameters for each dataset, FLix can accommodate diverse dataset mixtures and generalize better to unseen datasets. Our experiments show that FLix leads to significant improvements over a variety of tasks for both supervised learning and zero-shot settings using different training data mixtures.

Computation and Language

Acquiring Linguistic Knowledge from Multimodal Input

In contrast to children, language models (LMs) exhibit considerably inferior data efficiency when acquiring language. In this submission to the BabyLM Challenge (Warstadt et al., 2023), we test the hypothesis that this data efficiency gap is partly caused by a lack of multimodal input and grounding in the learning environment of typical language models. Although previous work looking into this question found that multimodal training can even harm language-only performance, we speculate that these findings can be attributed to catastrophic forgetting of complex language due to fine-tuning on captions data. To test our hypothesis, we perform an ablation study on FLAVA (Singh et al., 2022), a multimodal vision-and-language model, independently varying the volume of text and vision input to quantify how much text data (if any) can be offset by vision at different data scales. We aim to limit catastrophic forgetting through a multitask pretraining regime that includes unimodal text-only tasks and data sampled from WiT, the relatively diverse Wikipedia-based dataset (Srinivasan et al., 2021). Our results are largely negative: Multimodal pretraining does not harm our models' language performance but does not consistently help either. That said, our conclusions are limited by our having been able to conduct only a small number of runs. While we must leave open the possibility that multimodal input explains some of the gap in data efficiency between LMs and humans, positive evidence for this hypothesis will require better architectures and techniques for multimodal training.

Computation and Language

Large Language Models(LLMs) on Tabular Data: Prediction, Generation, and Understanding -- A Survey

Recent breakthroughs in large language modeling have facilitated rigorous exploration of their application in diverse tasks related to tabular data modeling, such as prediction, tabular data synthesis, question answering, and table understanding. Each task presents unique challenges and opportunities. However, there is currently a lack of comprehensive review that summarizes and compares the key techniques, metrics, datasets, models, and optimization approaches in this research domain. This survey aims to address this gap by consolidating recent progress in these areas, offering a thorough survey and taxonomy of the datasets, metrics, and methodologies utilized. It identifies strengths, limitations, unexplored territories, and gaps in the existing literature, while providing some insights for future research directions in this vital and rapidly evolving field. It also provides relevant code and datasets references. Through this comprehensive review, we hope to provide interested readers with pertinent references and insightful perspectives, empowering them with the necessary tools and knowledge to effectively navigate and address the prevailing challenges in the field.

Computation and Language

Gradient-Free Adaptive Global Pruning for Pre-trained Language Models

The transformative impact of large language models (LLMs) like LLaMA and GPT on natural language processing is countered by their prohibitive computational demands. Pruning has emerged as a pivotal compression strategy, introducing sparsity to enhance both memory and computational efficiency. Yet, traditional global pruning is impractical for LLMs due to scalability issues, while local pruning, despite its efficiency, leads to suboptimal solutions. Addressing these challenges, we propose Adaptive Global Pruning (AdaGP), a novel framework that redefines the global pruning process into manageable, coordinated subproblems, allowing for resource-efficient optimization with global optimality. AdaGP's approach, which conceptualizes LLMs as a chain of modular functions and leverages auxiliary variables for problem decomposition, not only facilitates a pragmatic application on LLMs but also demonstrates significant performance improvements, particularly in high-sparsity regimes where it surpasses current state-of-the-art methods.

Computation and Language

Multilingual Speech Models for Automatic Speech Recognition Exhibit Gender Performance Gaps

Current voice recognition approaches use multi-task, multilingual models for speech tasks like Automatic Speech Recognition (ASR) to make them applicable to many languages without substantial changes. However, broad language coverage can still mask performance gaps within languages, for example, across genders. We systematically evaluate multilingual ASR systems on gendered performance gaps. Using two popular models on three datasets in 19 languages across seven language families, we find clear gender disparities. However, the advantaged group varies between languages. While there are no significant differences across groups in phonetic variables (pitch, speaking rate, etc.), probing the model's internal states reveals a negative correlation between probe performance and the gendered performance gap. I.e., the easier to distinguish speaker gender in a language, the more the models favor female speakers. Our results show that group disparities remain unsolved despite great progress on multi-tasking and multilinguality. We provide first valuable insights for evaluating gender gaps in multilingual ASR systems. We release all code and artifacts at https://github.com/g8a9/multilingual-asr-gender-gap.

Computation and Language

An Iterative Associative Memory Model for Empathetic Response Generation

Empathetic response generation is to comprehend the cognitive and emotional states in dialogue utterances and generate proper responses. Psychological theories posit that comprehending emotional and cognitive states necessitates iteratively capturing and understanding associated words across dialogue utterances. However, existing approaches regard dialogue utterances as either a long sequence or independent utterances for comprehension, which are prone to overlook the associated words between them. To address this issue, we propose an Iterative Associative Memory Model (IAMM) for empathetic response generation. Specifically, we employ a novel second-order interaction attention mechanism to iteratively capture vital associated words between dialogue utterances and situations, dialogue history, and a memory module (for storing associated words), thereby accurately and nuancedly comprehending the utterances. We conduct experiments on the Empathetic-Dialogue dataset. Both automatic and human evaluations validate the efficacy of the model. Meanwhile, variant experiments on LLMs also demonstrate that attending to associated words improves empathetic comprehension and expression.

Computation and Language

Collaborative decoding of critical tokens for boosting factuality of large language models

The most common training pipeline for large language models includes pretraining, finetuning and aligning phases, with their respective resulting models, such as the pretrained model and the finetuned model. Finetuned and aligned models show improved abilities of instruction following and safe generation, however their abilities to stay factual about the world are impacted by the finetuning process. Furthermore, the common practice of using sampling during generation also increases chances of hallucination. In this work, we introduce a collaborative decoding framework to harness the high factuality within pretrained models through the concept of critical tokens. We first design a critical token classifier to decide which model to use for the next token, and subsequently generates the next token using different decoding strategies. Experiments with different models and datasets show that our decoding framework is able to reduce model hallucination significantly, showcasing the importance of the collaborative decoding framework.

Computation and Language

M3-VRD: Multimodal Multi-task Multi-teacher Visually-Rich Form Document Understanding

This paper presents a groundbreaking multimodal, multi-task, multi-teacher joint-grained knowledge distillation model for visually-rich form document understanding. The model is designed to leverage insights from both fine-grained and coarse-grained levels by facilitating a nuanced correlation between token and entity representations, addressing the complexities inherent in form documents. Additionally, we introduce new inter-grained and cross-grained loss functions to further refine diverse multi-teacher knowledge distillation transfer process, presenting distribution gaps and a harmonised understanding of form documents. Through a comprehensive evaluation across publicly available form document understanding datasets, our proposed model consistently outperforms existing baselines, showcasing its efficacy in handling the intricate structures and content of visually complex form documents.

Computation and Language

Exploring Multi-Document Information Consolidation for Scientific Sentiment Summarization

Modern natural language generation systems with LLMs exhibit the capability to generate a plausible summary of multiple documents; however, it is uncertain if models truly possess the ability of information consolidation to generate summaries, especially on those source documents with opinionated information. To make scientific sentiment summarization more grounded, we hypothesize that in peer review human meta-reviewers follow a three-layer framework of sentiment consolidation to write meta-reviews and it represents the logic of summarizing scientific sentiments in meta-review generation. The framework is validated via human annotation. Based on the framework, we propose evaluation metrics to assess the quality of generated meta-reviews, and we find that the hypothesis of the sentiment consolidation framework works out empirically when we incorporate it as prompts for LLMs to generate meta-reviews in extensive experiments.

Computation and Language

A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systems

This survey provides a comprehensive review of research on multi-turn dialogue systems, with a particular focus on multi-turn dialogue systems based on large language models (LLMs). This paper aims to (a) give a summary of existing LLMs and approaches for adapting LLMs to downstream tasks; (b) elaborate recent advances in multi-turn dialogue systems, covering both LLM-based open-domain dialogue (ODD) and task-oriented dialogue (TOD) systems, along with datasets and evaluation metrics; (c) discuss some future emphasis and recent research problems arising from the development of LLMs and the increasing demands on multi-turn dialogue systems.

Computation and Language

Hire a Linguist!: Learning Endangered Languages with In-Context Linguistic Descriptions

How can large language models (LLMs) process and translate endangered languages? Many languages lack a large corpus to train a decent LLM; therefore existing LLMs rarely perform well in unseen, endangered languages. On the contrary, we observe that 2000 endangered languages, though without a large corpus, have a grammar book or a dictionary. We propose LINGOLLM, a training-free approach to enable an LLM to process unseen languages that hardly occur in its pre-training. Our key insight is to demonstrate linguistic knowledge of an unseen language in an LLM's prompt, including a dictionary, a grammar book, and morphologically analyzed input text. We implement LINGOLLM on top of two models, GPT-4 and Mixtral, and evaluate their performance on 5 tasks across 8 endangered or low-resource languages. Our results show that LINGOLLM elevates translation capability from GPT-4's 0 to 10.5 BLEU for 10 language directions. Our findings demonstrate the tremendous value of linguistic knowledge in the age of LLMs for endangered languages. Our data, code, and model generations can be found at https://github.com/LLiLab/llm4endangeredlang.

Computation and Language

ResLoRA: Identity Residual Mapping in Low-Rank Adaption

As one of the most popular parameter-efficient fine-tuning (PEFT) methods, low-rank adaptation (LoRA) is commonly applied to fine-tune large language models (LLMs). However, updating the weights of LoRA blocks effectively and expeditiously is challenging due to the long calculation path in the original model. To address this, we propose ResLoRA, an improved framework of LoRA. By adding residual paths during training and using merging approaches to eliminate these extra paths during inference, our method can achieve better results in fewer training steps without any extra trainable parameters or inference cost compared to LoRA. The experiments on NLG, NLU, and text-to-image tasks demonstrate the effectiveness of our method. To the best of our knowledge, ResLoRA is the first work that combines the residual path with LoRA. The code of our method is available at https://github.com/microsoft/LMOps/tree/main/reslora .

Computation and Language

Datasets for Large Language Models: A Comprehensive Survey

This paper embarks on an exploration into the Large Language Model (LLM) datasets, which play a crucial role in the remarkable advancements of LLMs. The datasets serve as the foundational infrastructure analogous to a root system that sustains and nurtures the development of LLMs. Consequently, examination of these datasets emerges as a critical topic in research. In order to address the current lack of a comprehensive overview and thorough analysis of LLM datasets, and to gain insights into their current status and future trends, this survey consolidates and categorizes the fundamental aspects of LLM datasets from five perspectives: (1) Pre-training Corpora; (2) Instruction Fine-tuning Datasets; (3) Preference Datasets; (4) Evaluation Datasets; (5) Traditional Natural Language Processing (NLP) Datasets. The survey sheds light on the prevailing challenges and points out potential avenues for future investigation. Additionally, a comprehensive review of the existing available dataset resources is also provided, including statistics from 444 datasets, covering 8 language categories and spanning 32 domains. Information from 20 dimensions is incorporated into the dataset statistics. The total data size surveyed surpasses 774.5 TB for pre-training corpora and 700M instances for other datasets. We aim to present the entire landscape of LLM text datasets, serving as a comprehensive reference for researchers in this field and contributing to future studies. Related resources are available at: https://github.com/lmmlzn/Awesome-LLMs-Datasets.

Computation and Language

Crisis talk: analysis of the public debate around the energy crisis and cost of living

A prominent media topic in the UK in the early 2020s is the energy crisis affecting the UK and most of Europe. It brings into a single public debate issues of energy dependency and sustainability, fair distribution of economic burdens and cost of living, as well as climate change, risk, and sustainability. In this paper, we investigate the public discourse around the energy crisis and cost of living to identify how these pivotal and contradictory issues are reconciled in this debate and to identify which social actors are involved and the role they play. We analyse a document corpus retrieved from UK newspapers from January 2014 to March 2023. We apply a variety of natural language processing and data visualisation techniques to identify key topics, novel trends, critical social actors, and the role they play in the debate, along with the sentiment associated with those actors and topics. We combine automated techniques with manual discourse analysis to explore and validate the insights revealed in this study. The findings verify the utility of these techniques by providing a flexible and scalable pipeline for discourse analysis and providing critical insights for cost of living - energy crisis nexus research.

Computation and Language

Multi-FAct: Assessing Multilingual LLMs' Multi-Regional Knowledge using FActScore

Large Language Models (LLMs) are prone to factuality hallucination, generating text that contradicts established knowledge. While extensive research has addressed this in English, little is known about multilingual LLMs. This paper systematically evaluates multilingual LLMs' factual accuracy across languages and geographic regions. We introduce a novel pipeline for multilingual factuality evaluation, adapting FActScore(Min et al., 2023) for diverse languages. Our analysis across nine languages reveals that English consistently outperforms others in factual accuracy and quantity of generated facts. Furthermore, multilingual models demonstrate a bias towards factual information from Western continents. These findings highlight the need for improved multilingual factuality assessment and underscore geographical biases in LLMs' fact generation.

Computation and Language

Characterizing Truthfulness in Large Language Model Generations with Local Intrinsic Dimension

We study how to characterize and predict the truthfulness of texts generated from large language models (LLMs), which serves as a crucial step in building trust between humans and LLMs. Although several approaches based on entropy or verbalized uncertainty have been proposed to calibrate model predictions, these methods are often intractable, sensitive to hyperparameters, and less reliable when applied in generative tasks with LLMs. In this paper, we suggest investigating internal activations and quantifying LLM's truthfulness using the local intrinsic dimension (LID) of model activations. Through experiments on four question answering (QA) datasets, we demonstrate the effectiveness ohttps://info.arxiv.org/help/prep#abstractsf our proposed method. Additionally, we study intrinsic dimensions in LLMs and their relations with model layers, autoregressive language modeling, and the training of LLMs, revealing that intrinsic dimensions can be a powerful approach to understanding LLMs.

Computation and Language

MEGAnno+: A Human-LLM Collaborative Annotation System

Large language models (LLMs) can label data faster and cheaper than humans for various NLP tasks. Despite their prowess, LLMs may fall short in understanding of complex, sociocultural, or domain-specific context, potentially leading to incorrect annotations. Therefore, we advocate a collaborative approach where humans and LLMs work together to produce reliable and high-quality labels. We present MEGAnno+, a human-LLM collaborative annotation system that offers effective LLM agent and annotation management, convenient and robust LLM annotation, and exploratory verification of LLM labels by humans.

Computation and Language

Contextualizing Generated Citation Texts

Abstractive citation text generation is usually framed as an infilling task, where a sequence-to-sequence model is trained to generate a citation given a reference paper and the context window around the target; the generated citation should be a brief discussion of the reference paper as it relates to the citing context. However, examining a recent LED-based citation generation system, we find that many of the generated citations are generic summaries of the reference papers main contribution, ignoring the citation contexts focus on a different topic. To address this problem, we propose a simple modification to the citation text generation task: the generation target is not only the citation itself, but the entire context window, including the target citation. This approach can be easily applied to any abstractive citation generation system, and our experimental results show that training in this way is preferred by human readers and allows the generation model to make use of contextual clues about what topic to discuss and what stance to take.

Computation and Language

Benchmarking Large Language Models on Answering and Explaining Challenging Medical Questions

LLMs have demonstrated impressive performance in answering medical questions, such as passing scores on medical licensing examinations. However, medical board exam questions or general clinical questions do not capture the complexity of realistic clinical cases. Moreover, the lack of reference explanations means we cannot easily evaluate the reasoning of model decisions, a crucial component of supporting doctors in making complex medical decisions. To address these challenges, we construct two new datasets: JAMA Clinical Challenge and Medbullets. JAMA Clinical Challenge consists of questions based on challenging clinical cases, while Medbullets comprises USMLE Step 2&3 style clinical questions. Both datasets are structured as multiple-choice question-answering tasks, where each question is accompanied by an expert-written explanation. We evaluate four LLMs on the two datasets using various prompts. Experiments demonstrate that our datasets are harder than previous benchmarks. The inconsistency between automatic and human evaluations of model-generated explanations highlights the need to develop new metrics to support future research on explainable medical QA.

Computation and Language