OFA-Sys ORG

Official repository of OFA (ICML 2022). Paper: OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework

Chinese version of CLIP which achieves Chinese cross-modal retrieval and representation generation.

OFASys: A Multi-Modal Multi-Task Learning System for Building Generalist Models

A general representation model across vision, audio, language modalities. Paper: ONE-PEACE: Exploring One General Representation Model Toward Unlimited Modalities

An opensource ChatBot built with ExpertPrompting which achieves 96% of ChatGPT's capability.

InsTag: A Tool for Data Analysis in LLM Supervised Fine-tuning

Codes and Data for Scaling Relationship on Learning Mathematical Reasoning with Large Language Models

A self-ailgnment method for role-play. Benchmark for role-play. Resources for "Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment".

AIR-Bench: Benchmarking Large Audio-Language Models via Generative Comprehension

No description.

Self-Evolved Diverse Data Sampling for Efficient Instruction Tuning

Touchstone: Evaluating Vision-Language Models by Language Models

No description.

OFA-Compress is a unified framework which provides OFA model finetuning, distillation and inference capabilities in Huggingface version, and is committed to promoting the lightweighting of large models.

Pretrained Diffusion Models for Unified Human Motion Synthesis

present speech demo of FTTS