[ACL24] Official repo for "Iterative Forward Tuning Boosts In-Context Learning in Language Models"

Longitudinal Evaluation of LLMs via Data Compression

Lossless Training Speed Up by Unbiased Dynamic Data Pruning

1.4B sLLM for Chinese and English - HammerLLM🔨

Code for paper - On Diversified Preferences of Large Language Model Alignment

Code for Paper (ReMax: A Simple, Efficient and Effective Reinforcement Learning Method for Aligning Large Language Models)

Human preference data for "Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback"

Learning from synthetic data - code and models

Official github repo of G-LLaVA

This is the official repo for Contrastive Vision-Language Alignment Makes Efficient Instruction Learner.

📰 Must-read papers and blogs on LLM based Long Context Modeling 🔥

Dromedary: towards helpful, ethical and reliable LLMs.

✨✨Woodpecker: Hallucination Correction for Multimodal Large Language Models. The first work to correct hallucinations in MLLMs.

🦦 Otter, a multi-modal model based on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on MIMIC-IT and showcasing improved instruction-following and in-context learning ability.

MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities (ICML 2024)

MultimodalC4 is a multimodal extension of c4 that interleaves millions of images with text.

leaked prompts of GPTs

ACL 2024 | LooGLE: Long Context Evaluation for Long-Context Language Models

Automatically split your PyTorch models on multiple GPUs for training & inference

High Accuracy and efficiency multi-task fine-tuning framework for Code LLMs. This work has been accepted by KDD 2024.

code for Scaling Laws of RoPE-based Extrapolation

Reference implementation for DPO (Direct Preference Optimization)

A collection of GPT system prompts and various prompt injection/leaking knowledge.

Code and data for CoachLM, an automatic instruction revision approach LLM instruction tuning.

A dataset of LLM-generated chain-of-thought steps annotated with mistake location.

GPT-Fathom is an open-source and reproducible LLM evaluation suite, benchmarking 10+ leading open-source and closed-source LLMs as well as OpenAI's earlier models on 20+ curated benchmarks under al…