Train transformer language models with reinforcement learning.

Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.

RankLLM is a Python toolkit for reproducible information retrieval research using rerankers, with a focus on listwise reranking.

A lightweight, low-dependency, unified API to use all common reranking and cross-encoder models.

A machine learning benchmark of in-the-wild distribution shifts, with data loaders, evaluators, and default models.

Large Action Model framework to develop AI Web Agents

A dataset of atomic wikipedia edits containing insertions and deletions of a contiguous chunk of text in a sentence. This dataset contains ~43 million edits across 8 languages.

Dataset for Unified Editing, EMNLP 2023. This is a model editing dataset where edits are natural language phrases.

Reference implementation for DPO (Direct Preference Optimization)

[ICLR'24 spotlight] An open platform for training, serving, and evaluating large language model for tool learning.

Generative Representational Instruction Tuning

In-Context Learning for eXtreme Multi-Label Classification (XMC) using only a handful of examples.

Stanford NLP Python library for understanding and improving PyTorch models via interventions

Dataset of synthetic job ad sentences tagged with ESCO skills. From the paper Extreme Multi-Label Skill Extraction Training using Large Language Models.

A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).

Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages

Dataset used to evaluate Skill Extraction systems based on the ESCO skills taxonomy.

SKILLSPAN: Competences as Spans for Skill Extraction from Job Postings

The dataset used to evaluate JobBERT on the task of job title normalization.

TFIDF / KNN based string matching

State-of-the-art efficient coreference. This repository contains the code for the CRAC-2023 paper "CAW-coref: Conjunction-Aware Word-level Coreference Resolution". Forked from the EMNLP-2021 paper …

Inspecting and Editing Knowledge Representations in Language Models

BioDEX: Large-Scale Biomedical Adverse Drug Event Extraction for Real-World Pharmacovigilance.

OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.

High-speed download of LLaMA, Facebook's 65B parameter GPT model

QLoRA: Efficient Finetuning of Quantized LLMs

Home of StarCoder: fine-tuning & inference!

Public repo for the NeurIPS 2023 paper "Unlimiformer: Long-Range Transformers with Unlimited Length Input"

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Holistic Evaluation of Language Models (HELM) is an open source Python framework created by the Center for Research on Foundation Models (CRFM) at Stanford for holistic, reproducible and transparen…