Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
Paper
• 2504.13837 • Published
• 139
📝 Cool LLM papers
Starting from 2024-11-15
-
-
Understanding R1-Zero-Like Training: A Critical Perspective
Paper • 2503.20783 • Published • 59 -
Inference-Time Scaling for Generalist Reward Modeling
Paper • 2504.02495 • Published • 58 -
Large Language Diffusion Models
Paper • 2502.09992 • Published • 126 -
SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
Paper • 2502.02737 • Published • 255 -
WILDCHAT-50M: A Deep Dive Into the Role of Synthetic Data in Post-Training
Paper • 2501.18511 • Published • 20 -
Qwen2.5 Technical Report
Paper • 2412.15115 • Published • 377 -
SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language Models
Paper • 2412.11605 • Published • 18 -
Scaling test-time compute
📈593Boost LLM answers with search‑guided test‑time compute
-
Reverse Thinking Makes LLMs Stronger Reasoners
Paper • 2411.19865 • Published • 23 -
TÜLU 3: Pushing Frontiers in Open Language Model Post-Training
Paper • 2411.15124 • Published • 67 -
Scaling Laws for Precision
Paper • 2411.04330 • Published • 7 -
LoRA vs Full Fine-tuning: An Illusion of Equivalence
Paper • 2410.21228 • Published • 3 -
Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback
Paper • 2406.09279 • Published • 3 -
Preference Fine-Tuning of LLMs Should Leverage Suboptimal, On-Policy Data
Paper • 2404.14367 • Published • 1 -
Direct Language Model Alignment from Online AI Feedback
Paper • 2402.04792 • Published • 35 -
Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study
Paper • 2404.10719 • Published • 6 -
Extracting alignment data in open models
Paper • 2510.18554 • Published • 10