paper-with-me

Papers

GeneralThinker: Domain-General Reasoning through Likelihood-Guided Answer-Conditioned Optimization

2026-05-27 · Shengmin Piao, Sanghyun Park arxiv

Reinforcement learning with verifiable rewards improves language model reasoning, but its reliance on domain-specific verifiers, sparse outcome rewards, and coarse-grained credit assignment limits its applicability. We introduce GeneralThinker, an on-policy framework that reformulates reasoning supervision as dense answer-conditioned optimization, enabling response-level evaluation and token-level credit assignment without domain-specific verifiers. GeneralThinker evaluates generated reasoning trajectories using the likelihood of the ground-truth answer and derives token-wise compatibility signals for fine-grained credit assignment. To stabilize optimization, it constrains token-level updates through clipping and direction-preserving modulation. Across 11 benchmarks spanning mathematics, STEM, and general reasoning, GeneralThinker achieves the best average performance. Further analyses show that uncontrolled token-level modulation can destabilize training, whereas controlled modulation makes fine-grained credit assignment consistently effective.

📄 PDF Abstract BibTeX arXiv:2605.27934

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Towards Universal Neural Likelihood Inference

2025-08-12 · Shreyas Bhat Brahmavar, Yang Li, Qiyang Liu, Shashank Srivastava 외 arxiv

We introduce universal neural likelihood inference (UNLI): enabling a single model to provide data-grounded, conditional likelihood predictions for arbitrary targets given any collection of observed features, across dive…

All-Quadrant Bounded Clipping GRPO: Closing the Unbounded Blind Spot for Stable and Generalizable Training

2026-01-07 · Chi Liu, Xin Chen arxiv

Group Relative Policy Optimization (GRPO) has emerged as a popular algorithm for reinforcement learning with large language models (LLMs). However, GRPO inherits PPO's token-level clipping while replacing token-level adv…

Reinforcement LearningMathematical Reasoning

Capacity-Dependent Effects of Data Selection for Reasoning

2026-08-13 · Cuong Dang, Hoang Anh Just, Ruoxi Jia arxiv

In reasoning supervised fine-tuning, candidate responses for the same instruction can differ substantially in how well they match the student's current distribution. Recent likelihood-based response selection methods sug…

Mathematical Reasoning

Bonsai: Interpretable Tree-Adaptive Grounded Reasoning

2025-04-04 · Kate Sanders, Benjamin Van Durme

To develop general-purpose collaborative agents, humans need reliable AI systems that can (1) adapt to new domains and (2) transparently reason with uncertainty to allow for verification and correction. Black-box models …

Question AnsweringSpecificity

Bradley-Terry Policy Optimization for Generative Preference Modeling

2025-10-17 · Shengyu Feng, Yun He, Shuang Ma, Beibin Li 외 arxiv

Reinforcement learning (RL) has recently proven effective at scaling chain-of-thought (CoT) reasoning in large language models for tasks with verifiable answers. However, extending RL-based thought training to more gener…

Reinforcement Learning