Martin-qyma/TRM
From Faithfulness to Correctness: Generative Reward Models that Think Critically
This project offers a sophisticated tool for evaluating the quality of AI-generated answers in Chinese. It takes a question, an answer, and supporting documents as input, then critically assesses each sentence of the answer for how well it aligns with the documents and its factual accuracy. This is ideal for developers and researchers building or fine-tuning large language models for Chinese question answering.
No commits in the last 6 months.
Use this if you are developing or evaluating Chinese question-answering systems and need to rigorously assess the faithfulness and correctness of generated responses.
Not ideal if you are looking for a general-purpose language model or a tool for languages other than Chinese, or if you are not working on advanced AI model development.
Stars
14
Forks
1
Language
Python
License
—
Category
Last pushed
Oct 01, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/Martin-qyma/TRM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
agentscope-ai/Trinity-RFT
Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement...
OpenRLHF/OpenRLHF
An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO &...
zjunlp/EasyEdit
[ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs.
huggingface/alignment-handbook
Robust recipes to align language models with human and AI preferences
hyunwoongko/nanoRLHF
nanoRLHF: from-scratch journey into how LLMs and RLHF really work.