paper-with-me

홈 › Papers

Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models

2025-05-23 · Xuchen Pan, Yanxi Chen, Yushuo Chen, Yuchang Sun, Daoyuan Chen, WenHao Zhang, Yuexiang Xie, Yilun Huang, Yilei Zhang, Dawei Gao, Yaliang Li, Bolin Ding, Jingren Zhou

Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models. It is built with a decoupled design, consisting of (1) an RFT-core that unifies and generalizes synchronous/asynchronous, on-policy/off-policy, and online/offline modes of RFT, (2) seamless integration for agent-environment interaction with high efficiency and robustness, and (3) systematic data pipelines optimized for RFT. Trinity-RFT can be easily adapted for diverse application scenarios, and serves as a unified platform for exploring advanced reinforcement learning paradigms. This technical report outlines the vision, features, design and implementations of Trinity-RFT, accompanied by extensive examples demonstrating the utility and user-friendliness of the proposed framework.

📄 PDF Abstract BibTeX arXiv:2505.17826

Code (1)

modelscope/trinity-rft 공식 구현 pytorch

Similar Papers 제목 키워드 기반

TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems

2026-03-16 · Kai Wang, Biaojie Zeng, Zeming Wei, Chang Jin 외 arxiv

With the rapid development of LLM-based multi-agent systems (MAS), their significant safety and security concerns have emerged, which introduce novel risks going beyond single agents or LLMs. Despite attempts to address …

Too Helpful, Too Harmless, Too Honest or Just Right?

2025-09-10 · Gautam Siddharth Kashyap, Mark Dras, Usman Naseem arxiv

Large Language Models (LLMs) exhibit strong performance across a wide range of NLP tasks, yet aligning their outputs with the principles of Helpfulness, Harmlessness, and Honesty (HHH) remains a persistent challenge. Exi…

Trinity: Syncretizing Multi-/Long-tail/Long-term Interests All in One

2024-02-05 · Jing Yan, Liu Jiang, Jianfei Cui, Zhichen Zhao 외

Interest modeling in recommender system has been a constant topic for improving user experience, and typical interest modeling tasks (e.g. multi-interest, long-tail interest and long-term interest) have been investigated…

AllRecommendation Systems

The Trinity of Consistency as a Defining Principle for General World Models

2026-02-26 · Jingxuan Wei, Siyuan Li, Yuhang Xu, Zheng Sun 외 arxiv

The construction of World Models capable of learning, simulating, and reasoning about objective physical laws constitutes a foundational challenge in the pursuit of Artificial General Intelligence. Recent advancements re…

Video Generation

TRINITY: An Evolved LLM Coordinator

2025-12-04 · Jinglue Xu, Qi Sun, Peter Schwendeman, Stefan Nielsen 외 arxiv

Combining diverse foundation models is promising, but weight-merging is limited by mismatched architectures and closed APIs. Trinity addresses this with a lightweight coordinator that orchestrates collaboration among lar…

Reinforcement Learning