IDEAL: Influence-Driven Selective Annotations Empower In-Context Learners in Large Language Models

Examining how large language models (LLMs) perform across various synthetic regression tasks when given (input, output) examples in their context, without any parameter update

Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.

Does Refusal Training in LLMs Generalize to the Past Tense? [arXiv, July 2024]

Implementation of paper Data Engineering for Scaling Language Models to 128K Context

Leave One Out

An Open-source Framework for Autonomous Language Agents

An automatic evaluator for instruction-following language models. Human-validated, high-quality, cheap, and fast.

Official implementation for the paper *🎯DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving*

The official evaluation suite and dynamic data release for MixEval.

simple bibtex generator for any text with \cite{}

Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision

Self-Alignment with Principle-Following Reward Models

Lossless Training Speed Up by Unbiased Dynamic Data Pruning

For calculating Shapley values via linear regression.

The standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.

The interactive graphing library for Python ✨ This project now includes Plotly Express!

official code for "Large Language Models as Optimizers"

TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients.

Long Context Transfer from Language to Vision

A large-scale, fine-grained, diverse preference dataset (and models).

The test set for Koala

XLNet: Generalized Autoregressive Pretraining for Language Understanding