SocraticPO: Policy Optimization via Interactive Guidance

[2606.09887] SocraticPO: Policy Optimization via Interactive Guidance Skip to main content Search Submit Donate Log in Computer Science > Machine Learning arXiv:2606.09887 (cs) [Submitted on 3 Jun 2026 ( v1 ), last revised 22 Aug 2026 (this version, v2)] Title: SocraticPO: Policy Optimization via Interactive Guidance Authors: Zirui Liu , Tingyue Pan , Jie Ouyang , Qi Liu , Xianquan Wang , Jiayu Liu , Qingchuan Li , Jing Sha , Zhenya Huang , Shijin Wang , Enhong Chen View PDF HTML (experimental) Abstract: Reinforcement learning (RL) for large language models usually supervises reasoning with scalar outcome rewards, such as binary correctness. Such rewards provide an optimization direction but rarely explain how a model should revise its mistaken reasoning, which can encourage shortcut learning and brittle policies. We propose \textbf{SocraticPO} (Socratic Policy Optimization), a policy-optimization framework that augments RL rollouts with Socratic-style natural-language guidance. During rollout, the student first answers independently; if the answer is incorrect, a teacher diagnoses the attempt and provides concise corrective guidance, after which the student continues under the expanded context. Crucially, this guidance is paired with reward decay: correct answers obtained after teacher intervention only receive decayed rewards, preventing the policy from treating teacher help as a free path to reward. Since SocraticPO only modifies the rollout process while leaving the standard expected-reward objective intact, it can be plugged into existing policy-gradient backends such as Reinforce++. Moreover, because the teacher provides only text-level guidance, SocraticPO can leverage stronger black-box teacher models without requiring access to logits or distribution matching. On undergraduate-level scientific reasoning benchmarks from SciKnowEval, SocraticPO improves over strong RL and self-distillation baselines. Ablations show that both targeted guidance and reward decay are necessary, with reward decay mitigating reliance on assisted correction. Subjects: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2606.09887 [cs.LG] (or arXiv:2606.09887v2 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2606.09887 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Zirui Liu [ view email ] [v1] Wed, 3 Jun 2026 09:08:29 UTC (510 KB) [v2] Sat, 22 Aug 2026 17:46:25 UTC (511 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.LG < prev | next > new | recent | 2026-06 Change to browse by: cs cs.AI cs.CL 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? ) 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? ) 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

• SocraticPO: Policy Optimization via Interactive Guidance#665806
• Environment Parameter Gradient Theorem for Policy-Environment Co-Design in Reinforcement Learning#366272
• Fuel-optimal Boost-Back Guidance via Successive Convexification with a Terminal Constraint for ReusableLaunch Vehicles#1076415
• Constrained Group Relative Policy Optimization#924504
• Constrained Group Relative Policy Optimization#920273