LuckyyySTA

dify

Dify is an open-source LLM app development platform. Dify's intuitive interface combines AI workflow, RAG pipeline, agent capabilities, model management, observability features and more, letting you quickly go from prototype to production.

mem0

The Memory layer for your AI apps

graphrag

A modular graph-based Retrieval-Augmented Generation (RAG) system

doccano

Open source annotation tool for machine learning practitioners.

translation-agent

DeepSeek-Coder-V2

DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

chameleon

Repository for Meta Chameleon, a mixed-modal early-fusion foundation model from FAIR.

UMOE-Scaling-Unified-Multimodal-LLMs

The codes about "Uni-MoE: Scaling Unified Multimodal Models with Mixture of Experts"

bonito

A lightweight library for generating synthetic instruction tuning datasets for your data without GPT.

Awesome-Interpretability-in-Large-Language-Models

This repository collects all relevant resources about interpretability in LLMs

Step-DPO

Implementation for "Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs"

ToolkenGPT

ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings - NeurIPS 2023 (oral)

LLM-Tool-Survey

This is the repository for the Tool Learning survey.

AutoIF

loft

LOFT: A 1 Million+ Token Long-Context Benchmark

Humback

🐋 An unofficial implementation of Self-Alignment with Instruction Backtranslation.

retrieval-scaling

Official repository for "Scaling Retrieval-Based Langauge Models with a Trillion-Token Datastore".

Lookback-Lens

Official implementation for the paper "Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps"

DocGenome

DocGenome: An Open Large-scale Scientific Document Benchmark for Training and Testing Multi-modal Large Models

Internalize_CoT_Step_by_Step

swiftLLM

A tiny yet powerful LLM inference system tailored for researching purpose. vLLM-equivalent performance with only 2k lines of code (2% of vLLM).

LitSearch

A Retrieval Benchmark for Scientific Literature Search

RWKU

RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models

BRIGHT

BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrieval

llm_factuality_tuning

rag-qa-arena

llm_hallucinations

FactScoreLite

FactScoreLite is an implementation of the FactScore metric, designed for detailed accuracy assessment in text generation. This package builds upon the framework provided by the original FactScore repository, which is no longer maintained and contains outdated functions.

pragmatic_calibration

HalluPAQ

Leveraging Generated Q&A Pairs for Efficient Confidence Scoring and Hallucination Detection