Update README.md

README.md

@@ -6,4 +6,38 @@ inference:
     function_to_apply: "none"
 widget:
 - text: "I cuddled with my dog today."
----
+---
+
+# Conditional Utilitarian Deberta 01
+
+## Model description
+
+This is a [Deberta-based](https://huggingface.co/microsoft/deberta-v3-large) model. It was first fine-tuned on for computing utility estimates of experiences (see [utilitarian-deberta-01](https://huggingface.co/pfr/utilitarian-deberta-01). It was then further fine-tuned on 160 examples of conditional utilities pairwise comparisons.
+
+## Intended use
+
+The main use case is the computation of utility estimates of first-person text scenarios, under extra contextual information.
+
+## Limitations
+
+The model was fine-tuned on only 160 examples, so it should be expected to have limited performance.
+
+Further, while the base model was trained on ~10000 examples, they are still restricted, and only on first-person sentences. It does not have the capability of interpreting highly complex or unusual scenarios, and it does not have hard guarantees on its domain of accuracy. 
+
+## How to use
+
+Given a scenario S under a context C, and the model U, one computes the estimated conditional utility with `U(f'{C} {S}') - U(C)`.
+
+## Training data
+
+The first training data is the train split from the Utilitarianism part of the [ETHICS dataset](https://arxiv.org/abs/2008.02275).
+
+The second training data consists of 160 crowdsourced examples of triples (S, C0, C1) consisting of one scenario and two possible contexts, where `U(S | C0) > U(S | C1)`.
+
+## Training procedure
+
+Starting from [utilitarian-deberta-01](https://huggingface.co/pfr/utilitarian-deberta-01), we fine-tune the model over the training data of 160 examples, with a learning rate of `1e-5`, a batch size of `8`, and `nepochs` of 2.
+
+## Evaluation results
+
+The model achieves ~80% accuracy over 40 crowdsourced examples, from the same distribution as the training data.