Learning from Hard Prompts: Difficulty-aware Advantage Amplification in Dynamic Sampling

[2608.27982] Learning from Hard Prompts: Difficulty-aware Advantage Amplification in Dynamic Sampling Skip to main content Search Submit Donate Log in Computer Science > Artificial Intelligence arXiv:2608.27982 (cs) [Submitted on 28 Aug 2026] Title: Learning from Hard Prompts: Difficulty-aware Advantage Amplification in Dynamic Sampling Authors: Siyuan Gan , Yuhan Li , Xiran Wang , Linjian Meng , Boyan Wang , Zhen Zhao , Jing Huo , Lei Bai , Yang Gao View PDF HTML (experimental) Abstract: Decoupled Clip and Dynamic Sampling Policy Optimization (DAPO) is a prominent variant of Group Relative Policy Optimization (GRPO). DAPO introduces several improvements over GRPO. Among these, Dynamic Sampling contributes the most to DAPO's accuracy gains relative to GRPO. To improve accuracy, Dynamic Sampling enhances training stability by eliminating zero policy gradients from zero advantages. Specifically, it avoids such zero gradients by filtering out prompts where sampled responses are either entirely correct or incorrect. However, our theoretical analysis shows that Dynamic Sampling decrease training efficiency as it cannot effectively utilize hard-to-sample correct responses on hard prompts. Formally, it asymmetrically amplifies the advantages of distinct responses to the same prompts. On hard prompts, incorrect responses undergo greater amplification than correct ones. This leads the model to avoid generating the observed incorrect responses rather than capitalizing on the hard-to-sample correct ones on hard prompts, resulting in low training efficiency. To improve training efficiency, we propose Direct Advantage Amplification (DAA), which amplifies the advantages of hard-to-sample correct responses on hard prompts, as obtained by Dynamic Sampling. This ensures that, when Dynamic Sampling is used, these hard-to-sample responses can be effectively capitalized on, implying higher training efficiency. By integrating DAA into DAPO, we obtain Difficulty-aware Advantage Amplification Policy Optimization (DA3PO), which is implemented with fewer than 30 lines of code from DAPO. Experiments show that DA3PO significantly outperforms GRPO and other classical GRPO variants. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.27982 [cs.AI] (or arXiv:2608.27982v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.27982 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Siyuan Gan [ view email ] [v1] Fri, 28 Aug 2026 06:51:31 UTC (73 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.AI < prev | next > new | recent | 2026-08 Change to browse by: cs References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation 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? ) 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? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)

Related documents

• Learning from Hard Prompts: Difficulty-aware Advantage Amplification in Dynamic Sampling#783167
• Clustering-Based Balanced Sampling and Allocation with Data Parallelism for High-Performance Fine-Tuning#1070074
• Verifying Sampling Algorithms via Distributional Invariants / Daniel Zilken, Kevin Batz, Joost-Pieter Katoen, Tobias Winkler#780428
• GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning#130757
• Towards Scalable RLVR: Multimodal Instruction Following Data Synthesis and Distillation#1110451