An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & RingAttention & RFT)

Code for paper "Unraveling Cross-Modality Knowledge Conflicts in Large Vision-Language Models."

A flexible and efficient training framework for large-scale alignment tasks

Repository for "Granular Privacy Control for Geolocation with Vision Language Models"

Python package for scraping real estate property data

Public code repo for paper "SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationales"

LLaMA 2 implemented from scratch in PyTorch

[ECCV'24] Official Implementation of Autoregressive Visual Entity Recognizer.

Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

Code for the ICML 2021 (long talk) paper: "ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision"

An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.

Official Repository for Can Language Models be Instructed to Protect Personal Information?

A collection of AWESOME things about mixture-of-experts

Code and Data for using the MultiSim Benchmark

EMNLP2023 - InfoSeek: A New VQA Benchmark focus on Visual Info-Seeking Questions

ICCV 2023 (Oral) Open-domain Visual Entity Recognition Towards Recognizing Millions of Wikipedia Entities

EMNLP 2020 GigaBERT Arabic Relation extraction system, named entity recognition, IE

EMNLP2021 Model Selection for Cross-Lingual Transfer

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

Recent advancements propelled by large language models (LLMs), encompassing an array of domains including Vision, Audio, Agent, Robotics, Fundamental Sciences such as Mathematics, and Ominous.

Build, evaluate, understand, and fix LLM-based apps

ACL 2023 (Findings) End-to-end Cross-lingual Label Project

Translation and Annotation Fusion for Cross-lingual transfer in low-resource languages

This reporsitory contains metadata of WavCaps dataset and codes for downstream tasks.

Mapping Wikipedia pages to Wikidata IDs and vice versa.