paper-with-me

Papers

LambdaPO: A Lambda Style Policy Optimization for Reasoning Language Models

2026-05-19 · Redacted by arXiv arxiv

Group Relative Policy Optimization(GRPO) has become a cornerstone of modern reinforcement learning alignment, prized for its efficacy in foregoing an explicit value-critic by leveraging reward normalization across sampled trajectory cohorts. However, the method's reliance on a monolithic statistical baseline, such as the group mean, collapses the relational topology of the trajectory space into a single scalar, thereby erasing the fine-grained preference information essential for navigating complex, rank-sensitive reward landscapes. To address this issue, we introduce a novel framework, Lambda Policy Optimization (LambdaPO), that addresses this information-theoretic bottleneck by re-conceptualizing advantage estimation from a scalar value to a decomposed, pairwise preference structure. Specifically, the advantage for any given trajectory is formulated as the integrated sum of reward differentials against all peers in its cohort, where each pairwise comparison is dynamically attenuated by the policy's own probabilistic confidence in the established preference. To further mitigate the sparsity of binary outcome supervision, we augment the objective with a semantic density reward, derived from the precision-recall alignment between generated reasoning traces and ground-truth solutions. As a result, our method can mine more fine-grained optimization signals from a group of rollouts, guiding the LLM to a better optima. Experimental results across challenging math reasoning and question-answering tasks demonstrates that LambdaPO improves performance compared to the baseline methods.

📄 PDF Abstract BibTeX arXiv:2605.19416

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

LamPO: A Lambda Style Policy Optimization for Reasoning Language Models

2026-05-20 · Redacted by arXiv arxiv

Reinforcement learning with verifiable rewards (RLVR) has become an effective paradigm for improving reasoning language models on tasks such as mathematics, coding, and scientific question answering. However, widely used…

Reinforcement LearningQuestion Answering

An Adaptive Clipping Approach for Proximal Policy Optimization

2018-04-17 · Gang Chen, Yiming Peng, Mengjie Zhang

Very recently proximal policy optimization (PPO) algorithms have been proposed as first-order optimization methods for effective reinforcement learning. While PPO is inspired by the same learning theory that justifies tr…

Learning TheoryReinforcement Learning

Off-policy Learning with Eligibility Traces: A Survey

2013-04-15 · Matthieu Geist, Bruno Scherrer

In the framework of Markov Decision Processes, off-policy learning, that is the problem of learning a linear approximation of the value function of some fixed policy from one trajectory possibly generated by some other p…

Survey

Constrained Policy Optimization via Bayesian World Models

2022-01-24 · ICLR 2022 4 · Yarden As, Ilnura Usmanova, Sebastian Curi, Andreas Krause

Improving sample-efficiency and safety are crucial challenges when deploying reinforcement learning in high-stakes real world applications. We propose LAMBDA, a novel model-based approach for policy optimization in safet…

reinforcement-learningReinforcement Learning (RL)

Limits of n-gram Style Control for LLMs via Logit-Space Injection

2026-01-12 · Sami-ul Ahmed arxiv

Large language models (LLMs) are typically personalized via prompt engineering or parameter-efficient fine-tuning such as LoRA. However, writing style can be difficult to distill into a single prompt, and LoRA fine-tunin…

parameter-efficient fine-tuningPrompt Engineering