paper-with-me

Papers

Boosting Inference with Guided Reasoning: Stochastic Exploration for Recursive Models

2026-05-24 · Andrew Corbett, Archit Sood, Anna Tzatzopoulou, Sai-Aakash Ramesh, Tim Dodwell arxiv

Recent work on recursive architectures has shown that tiny neural networks can be surprisingly powerful on structured reasoning tasks. The trick is to model reasoning trajectories with a latent dynamical system. We argue that the inference-time behaviour of these architectures is best understood as approximate inference over latent reasoning trajectories, with deterministic recursion as the one-particle, zero-noise limit. We make this view operational through guided stochastic exploration: stochastic perturbations of the reasoning dynamics propose neighbouring trajectories, and the model's existing early-stopping head reweights them online. The framework yields three label-free diagnostics: local stability, guide alignment, and cloud-token entropy. These predict, from inference traces alone, whether the procedure will help and which of its outputs to trust. On Sudoku-Extreme it lifts exact-solve accuracy from $85.9\%$ to $98.0\%$ without retraining; on Maze-Hard the diagnostics flag a misaligned guide, as validation performance later confirms. The same machinery thus characterises both when recursive reasoning has room to improve at the trajectory level and when the model's internal guide can recover it.

📄 PDF Abstract BibTeX arXiv:2605.25230

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

From Implicit Exploration to Structured Reasoning: Leveraging Guideline and Refinement for LLMs

2025-09-08 · Jiaxiang Chen, Zhuo Wang, Mingxi Zou, Zhucong Li 외 arxiv

Large language models (LLMs) have advanced general-purpose reasoning, showing strong performance across diverse tasks. However, existing methods often rely on implicit exploration, where the model follows stochastic and …

Reasoning While Recommending: Entropy-Guided Latent Reasoning in Generative Re-ranking Models

2026-01-20 · Changshuo Zhang arxiv

Reinforcement learning plays a crucial role in generative re-ranking scenarios due to its exploration-exploitation capabilities, but existing generative methods mostly fail to adapt to the dynamic entropy changes in mode…

Reinforcement Learning

Beyond Stochastic Exploration: What Makes Training Data Valuable for Agentic Search

2026-04-09 · Chuzhan Hao, Wenfeng Feng, Guochao Jiang, Guofeng Quan 외 arxiv

Reinforcement learning (RL) has become an effective approach for advancing the reasoning capabilities of large language models (LLMs) through the strategic integration of external search engines. However, current RL-base…

Reinforcement LearningMathematical Reasoning

Collaboratively boosting data-driven deep learning and knowledge-guided ontological reasoning for semantic segmentation of remote sensing imagery

2020-10-06 · Yansheng Li, Song Ouyang, Yongjun Zhang

As one kind of architecture from the deep learning family, deep semantic segmentation network (DSSN) achieves a certain degree of success on the semantic segmentation task and obviously outperforms the traditional method…

Segmentation Of Remote Sensing ImagerySemantic Segmentation

Reasoning Palette: Modulating Reasoning via Latent Contextualization for Controllable Exploration for (V)LMs

2025-12-19 · Rujiao Long, Yang Li, Xingyao Zhang, Weixun Wang 외 arxiv

Exploration capacity shapes both inference-time performance and reinforcement learning (RL) training for large (vision-) language models, as stochastic sampling often yields redundant reasoning paths with little high-lev…

Reinforcement Learning