paper-with-me

홈 › Papers

Beyond Trajectory Imitation: Strategy-Guided Policy Optimization for LLM Reasoning

2026-06-23 · Tianyuan Shi, Canbin Huang, Bei Li, Xin Chen, Xiaojun Quan, Jingang Wang, Qifan Wang arxiv

Distilling reasoning capabilities from strong to weak language models typically involves imitating specific solution trajectories, effectively transferring what to answer rather than how to reason. This trajectory-level imitation encourages memorization of instance-specific steps rather than acquisition of transferable problem-solving skills, limiting generalization to novel problems. We propose Strategy-Guided Policy Optimization (SGPO), which replaces instance-level trajectory imitation with reusable strategy distillation. SGPO extracts structured strategy descriptions from strong-model responses and, for each problem, constructs both autonomous and strategy-guided trajectories to enable direct comparison of the model's behavior with and without strategic guidance. The framework then addresses two key questions. For how to distill, a token-level forward-KL objective selectively transfers the distributional shift induced by strategy conditioning into the unguided policy, with proximal constraints ensuring stability. For when to distill, adaptive instance-level weighting strengthens guidance when autonomous exploration falls short and reduces it as the model's own competence grows. Experiments on four mathematical benchmarks across two model families show that SGPO consistently outperforms SFT, on-policy RL, and hybrid-policy baselines, improving the average score by 2.2 points over the strongest baseline on Qwen2.5-7B-Instruct. Analysis reveals that the forward-KL objective provides an inherently selective distillation signal that outperforms direct trajectory imitation, and that strategy distillation exhibits complementary scaling with base model capability.

📄 PDF Abstract BibTeX arXiv:2606.24064

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Cross-Domain Energy-Guided Diffusion Generation for Off-Dynamics Reinforcement Learning

2026-05-24 · Yu Yang, Yihong Guo, Anqi Liu, Pan Xu arxiv

Off-dynamics offline reinforcement learning seeks to learn a target-domain policy from a large source dataset and a limited target dataset under mismatched transition dynamics. Existing approaches such as reward augmenta…

Reinforcement LearningOffline RL

Sketch-to-Skill: Bootstrapping Robot Learning with Human Drawn Trajectory Sketches

2025-03-14 · Peihong Yu, Amisha Bhaskar, Anukriti Singh, Zahiruddin Mahammad 외

Training robotic manipulation policies traditionally requires numerous demonstrations and/or environmental rollouts. While recent Imitation Learning (IL) and Reinforcement Learning (RL) methods have reduced the number of…

Imitation Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

CACTO: Continuous Actor-Critic with Trajectory Optimization -- Towards global optimality

2022-11-12 · Gianluigi Grandesso, Elisa Alboni, Gastone P. Rosati Papini, Patrick M. Wensing 외

This paper presents a novel algorithm for the continuous control of dynamical systems that combines Trajectory Optimization (TO) and Reinforcement Learning (RL) in a single framework. The motivations behind this algorith…

continuous-controlContinuous ControlReinforcement Learning (RL)

Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR

2026-03-27 · Jinda Lu, Junkang Wu, Jinghan Li, Kexin Huang 외 arxiv

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) for multimodal large language models (MLLMs) have mainly focused on improving final answer correctness and strengthening visual grounding. However,…

Reinforcement LearningMultimodal ReasoningLogical ReasoningVisual Grounding

Policy-Based Trajectory Clustering in Offline Reinforcement Learning

2025-06-10 · Hao Hu, Xinqi Wang, Simon Shaolei Du

We introduce a novel task of clustering trajectories from offline reinforcement learning (RL) datasets, where each cluster center represents the policy that generated its trajectories. By leveraging the connection betwee…

ClusteringD4RLOffline RLreinforcement-learning+4