Laos_System provides a configurable end-to-end pipeline that converts clinical speech/text notes into structured JSON documents for: admission, surgery and discharge.

Complex Reasoning Rag System, Agentic Rag System

Complete EAGLE3-based speculative decoding solution for Kimi-K2-Instruct, including the draft model and training dataset, achieving a 3+ token acceptance rate on HumanEval, GSM8K, Math500, etc.

MrlX: A Multi-Agent Reinforcement Learning Framework

RAG-QA Leaderboard

[AAAI 2026]RAG, Benchmark, robust RAG generation

Medical Hallucination Benchmark

The MedEQBench is the first evaluation suite specifically designed for medical contexts to assess Large Language Models' (LLMs) capabilities in emotional perception (recognition/understanding) and empathic expression (supportive, safe, and context-aware responses).

A comprehensive medical AI evaluation framework based on GAPS methodology. Features automated assessment pipeline, thoracic surgery dataset (92 cases), multi-model support, parallel processing, and clinical scoring. Supports OpenAI-compatible APIs including GPT, Claude, Gemini, Qwen with detailed analytics and visualization.

[QReward] RewardService Python Client, make RL Training reward function more faster