Image de couverture de livre

BASIS: Batchwise Advantage Estimation from Single-Rollout Information Sharing for LLM Reasoning

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Gong, Shijin, Xu, Erhan, Ye, Kai, Quinzan, Francesco, Livieri, Giulia, Shi, Chengchun
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Kernelized Advantage Estimation: From Nonparametric Statistics to LLM Reasoning
par: Gong, Shijin, et autres
Publié: (2026)

Doubly Robust Alignment for Large Language Models
par: Xu, Erhan, et autres
Publié: (2025)

Demystifying Group Relative Policy Optimization: Its Policy Gradient is a U-Statistic
par: Zhou, Hongyi, et autres
Publié: (2026)

Robust Reinforcement Learning from Human Feedback for Large Language Models Fine-Tuning
par: Ye, Kai, et autres
Publié: (2025)

Learn-to-Distance: Distance Learning for Detecting LLM-Generated Text
par: Zhou, Hongyi, et autres
Publié: (2026)

Statistical Guarantees for Reasoning Probes on Looped Boolean Circuits
par: Kratsios, Anastasis, et autres
Publié: (2026)

Designing Universal Causal Deep Learning Models: The Case of Infinite-Dimensional Dynamical Systems from Stochastic Analysis
par: Galimberti, Luca, et autres
Publié: (2022)

Statistical Inference in Reinforcement Learning: A Selective Survey
par: Shi, Chengchun
Publié: (2025)

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning
par: Wu, Junkang, et autres
Publié: (2025)

How to Allocate, How to Learn? Dynamic Rollout Allocation and Advantage Modulation for Policy Optimization
par: Fang, Yangyi, et autres
Publié: (2026)

Learning Decision Policies with Instrumental Variables through Double Machine Learning
par: Shao, Daqian, et autres
Publié: (2024)

Detecting Changes in Causal Dependence with Kernels and Copulas
par: Gavioli-Akilagun, Shakeel, et autres
Publié: (2026)

Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning
par: Xu, Yixuan Even, et autres
Publié: (2025)

Double Machine Learning for Conditional Moment Restrictions: IV Regression, Proximal Causal Learning and Beyond
par: Shao, Daqian, et autres
Publié: (2025)

Segmental Advantage Estimation: Enhancing PPO for Long-Context LLM Training
par: Gong, Xue, et autres
Publié: (2026)

Pruning Cannot Hurt Robustness: Certified Trade-offs in Reinforcement Learning
par: Pedley, James, et autres
Publié: (2025)

Single-Rollout Hidden-State Dynamics for Training-Free RLVR Data Selection
par: Wu, Jianghao, et autres
Publié: (2026)

Stable and Efficient Single-Rollout RL for Multimodal Reasoning
par: Liu, Rui, et autres
Publié: (2025)

Controlling Transient Amplification Improves Long-horizon Rollouts
par: Pervez, Adeel, et autres
Publié: (2026)

From Authors to Reviewers: Leveraging Rankings to Improve Peer Review
par: Wang, Weichen, et autres
Publié: (2025)

AI Alignment in Medical Imaging: Unveiling Hidden Biases Through Counterfactual Analysis
par: Ma, Haroui, et autres
Publié: (2025)

Double Machine Learning Based Structure Identification from Temporal Data
par: Angelis, Emmanouil, et autres
Publié: (2023)

Learning Counterfactually Invariant Predictors
par: Quinzan, Francesco, et autres
Publié: (2022)

PEARL: Performance-Enhanced Aggregated Representation Learning
par: Li, Wenhui, et autres
Publié: (2025)

SHAPE: Stage-aware Hierarchical Advantage via Potential Estimation for LLM Reasoning
par: Ai, Zhengyang, et autres
Publié: (2026)

Detecting LLM-Generated Text with Performance Guarantees
par: Zhou, Hongyi, et autres
Publié: (2026)

Optimistic Model Rollouts for Pessimistic Offline Policy Optimization
par: Zhai, Yuanzhao, et autres
Publié: (2024)

Accelerating RL for LLM Reasoning with Optimal Advantage Regression
par: Brantley, Kianté, et autres
Publié: (2025)

Reinforcement Learning from Human Feedback: A Statistical Perspective
par: Liu, Pangpang, et autres
Publié: (2026)

Counterfactually Safe Reinforcement Learning
par: Li, Jingyi, et autres
Publié: (2026)

Dual Active Learning for Reinforcement Learning from Human Feedback
par: Liu, Pangpang, et autres
Publié: (2024)

Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts
par: Zheng, Haizhong, et autres
Publié: (2025)

Deep Generative Demand Learning for Newsvendor and Pricing
par: Gong, Shijin, et autres
Publié: (2024)

Neural Aided Kalman Filtering for UAV State Estimation in Degraded Sensing Environments
par: Gupta, Akhil, et autres
Publié: (2026)

Learning from one graph: transductive learning guarantees via the geometry of small random worlds
par: Detering, Nils, et autres
Publié: (2025)

AAPO: Enhancing the Reasoning Capabilities of LLMs with Advantage Margin
par: Xiong, Jian, et autres
Publié: (2025)

Learning Perturbations to Extrapolate Your LLM
par: Cen, Zetai, et autres
Publié: (2026)

DualKV: Shared-Prompt Flash Attention for Efficient RL Training with Large Rollouts and Long Contexts
par: Gai, Jiading, et autres
Publié: (2026)

BREAD: Branched Rollouts from Expert Anchors Bridge SFT & RL for Reasoning
par: Zhang, Xuechen, et autres
Publié: (2025)

Representation Invariance and Allocation: When Subgroup Balance Matters
par: Alloula, Anissa, et autres
Publié: (2025)