DRIFT大模型自進化框架發佈

Computer Science > Machine Learning arXiv:2606.

30345 (cs) [Submitted on 29 Jun 2026 (v1), last revised 10 Aug 2026 (this version, v4)] Title:DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training Authors:Haisen Luo, Yiwei Liu, Haoning Wang, Dan Liu, Junxi Yin, Haotian Wang, Lei Zhang, Xiaoyu Tian, Shuaiting Chen, Yuansheng Song, Baoyan Guo, Xiongfei Yan, Bolan Yang, Chengwei Liu, Ming Cui, Jiong Chen View a PDF of the paper titled DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training, by Haisen Luo and 15 other authors View PDF HTML (experimental) Abstract:Enabling large language models to achieve stable self-improvement without external expert supervision remains a central challenge in complex reasoning tasks.

Existing self-distillation and reinforcement learning methods lack explicit mechanisms for tracking problem-level learning progress and adapting optimization strategies accordingly.

Consequently, training may over-optimize easy problems, receive weak supervision from hard problems, and fail to sufficiently explore borderline cases.To resolve these issues, we propose DRIFT, an online self-evolution policy optimization framework for large language models.

DRIFT regulates the model's self-improvement process through the joint use of Difficulty Routing and Rhythm Gating.

The former identifies the model's learning state at the problem level and dynamically allocates self-distillation and reinforcement learning signals, while the latter refines policy updates at the token level, concentrating exploration on critical reasoning positions.

By further incorporating a success buffer and a two-stage curriculum learning strategy, DRIFT preserves high-quality historical experience while progressively guiding the model from reliable behavior acquisition toward stable policy evolution.

Evaluated across five benchmarks and three model scales, DRIFT surpasses the peak performance of both GRPO and SDPO across all evaluated metrics.On the average score over the five benchmarks, DRIFT achieves 79.5$\%$, outperforming GRPO by 9.5$\%$ and SDPO by 7.

5$\%$, establishing a new state-of-the-art result.Notably, on ToolUse, DRIFT reaches an accuracy of 79.2$\%$, improving over GRPO by 13.5$\%$ and SDPO by 10.7$\%$, setting a new state-of-the-art and substantially outperforming all concurrent methods.Subjects: Machine Learning (cs.

LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2606.30345 [cs.LG] (or arXiv:2606.30345v4 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2606.

30345 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Haoning Wang [view email] [v1] Mon, 29 Jun 2026 14:20:47 UTC (5,472 KB) [v2] Fri, 31 Jul 2026 04:05:32 UTC (5,574 KB) [v3] Mon, 3 Aug 2026 04:06:03 UTC (5,574 KB) [v4] Mon, 10 Aug 2026 05:19:37 UTC (5,574 KB) Full-text links: Access Paper: View a PDF of the paper titled DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training, by Haisen Luo and 15 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.

LG < prev | next > new | recent | 2026-06 Change to browse by: cs cs.AI References & Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading...BibTeX formatted citation × loading...

Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.

ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?

) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.

AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?

) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy.arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community?Learn more about arXivLabs.Which authors of this paper are endorsers?| Disable MathJax (What is MathJax?)