Reinforcement learning

Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.

6,937 stories · RSS feed

arXiv AI

Sep 1

CDPR: Counterfactual Advantage-based Credit Assignment for Cost-Aware Sequential Medical Diagnosis

arXiv:2608.28599v1 Announce Type: new Abstract: Clinical diagnosis is a step-by-step, cost-aware process: a physician orders examinations one at a time, observes the results, and updates the diagnosi...

By Qi Peng, Yi Cai, Changmeng Zheng, Xin Wu, Jiayuan Xie, Qing Li

llms reinforcement-learning benchmarks

arXiv AI

Sep 1

When Should Models Change Their Minds? Contextual Belief Management in Large Language Models

arXiv:2605.30219v2 Announce Type: replace Abstract: Long-horizon interactions require language models to manage accumulating information: when to update their state, when to preserve their state, and...

By Haoming Xu, Weihong Xu, Zongrui Li, Mengru Wang, Yunzhi Yao, Chiyu Wu, Jin Shang, Yu Gong, Shumin Deng

llms reinforcement-learning benchmarks

arXiv AI

Sep 1

SemPOI-RL: Aligning LLM Semantic Reasoning for Interpretable Out-of-Town POI Sequential Generation

arXiv:2608.30399v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit strong semantic reasoning and open-ended generation abilities, but aligning these abilities with structured sequ...

By Yunqi Liu, Yang Zhang, Ruixing Zhang, Liangzhe Han, Yi Qiao, Tongyu Zhu, Leilei Sun

llms reinforcement-learning fine-tuning safety

arXiv AI

Sep 1

How do World Models and Policies Compose in LLM Agents? A Joint Spectral and Behavioral Account

arXiv:2608.30067v1 Announce Type: cross Abstract: How do LLM agents come to both understand environments they act in and master tasks set within them? Through controlled experiments combining world-m...

By Ruize Xu, Xiao Yu, Yujin Tang, Chenming Shang, Nikhil Singh

llms agents reinforcement-learning

arXiv AI

Sep 1

Aligning Multi-Trajectory Supervision with Policy Optimization for VLA Driving

arXiv:2608.30122v1 Announce Type: cross Abstract: Vision-language-action (VLA) driving methods increasingly combine multi-trajectory imitation learning with group-relative policy optimization (GRPO),...

By Tian Zhang, Zhuo Huang, Hongrui Ye, Yu Wu, Zengmao Wang, Kaixuan Zhou

reinforcement-learning efficiency multimodal safety

arXiv AI

Sep 1

Hybrid Offline-Online Multi-Agent Decision Transformers for Wireless Resource Management

arXiv:2608.28878v1 Announce Type: cross Abstract: This paper develops a hybrid offline-online multi-agent reinforcement learning framework based on decision transformers. The policy is first pretrain...

By Yiming Zhang, Kun Yang, Cong Shen, Dongning Guo

llms agents reinforcement-learning fine-tuning

arXiv AI

Sep 1

MedAgent-R1: Faithfulness-Aware Reinforcement Learning for Evidence-Grounded Medical Reasoning

arXiv:2608.30676v1 Announce Type: new Abstract: When medical AI systems hallucinate clinical reasoning, the consequences extend beyond incorrect answers: fabricated justifications that superficially...

By Jiangwang Chen, Chenghao Zhang, Hengxing Cai

llms agents reinforcement-learning safety

arXiv AI

Sep 1

TRACER: Per-Tool Context Retention for LLM Agents via Consequence-Attributed Reinforcement Learning

arXiv:2608.29363v1 Announce Type: new Abstract: Enterprise data agents answer business queries by chaining many tool calls over multiple reasoning steps, routinely accumulating hundreds of thousands...

By Ziqi Lin, Ye Wu, Mengying Yang, Xu Liu, Yizhou Liu, Qiang Ke, Qin Guo

llms agents reinforcement-learning

arXiv Computation and Language

Sep 1

When History Is Multimodal: Rethinking Context Management for Long-Horizon Agents

arXiv:2608.29897v1 Announce Type: new Abstract: Long-horizon agents need a context manager to compress growing interaction histories into a bounded working context, via passive strategies or active s...

By Jiaqi Su, Cong Pang, Jiawei Hong, Tiankuo Yao, Zixuan Chen, Xin Lou, Lewei Lu

agents nlp reinforcement-learning fine-tuning efficiency multimodal benchmarks

arXiv AI

Sep 1

Cross-Relational Preference Learning for Better LLM Instruction Following

arXiv:2608.29352v1 Announce Type: new Abstract: Large Language Models (LLMs) still exhibit limited capability in following complex instructions. While existing approaches often rely on preference lea...

By Runsheng Li, Kai Sun, Bin Shi, Bo Dong

llms reinforcement-learning benchmarks

arXiv AI

Sep 1

StageWell: A Process-Aligned Chinese Corpus for Positive-Psychology Support Dialogue

arXiv:2608.29326v1 Announce Type: cross Abstract: Positive psychology dialogue aims to support emotional distress and positive resource building, requiring models to produce not only empathetic repli...

By Yuxiong Wang, Ziwei Lin, Bo Wang, Yu Zhang, Shiguang Ni

llms agents reinforcement-learning fine-tuning safety

arXiv Computation and Language

Sep 1

Mechanistic Diagnostics of Spatial Lexical Bias in Multimodal Large Language Model Spatial Reasoning

arXiv:2606.01914v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) remain unreliable on spatial multiple-choice questions, and their failures are often attributed to poorly...

By Chuang Ma, Qianying Liu, Tomoyuki Obuchi, Fei Cheng, Wang Yang, Sudong Cai, Shuyuan Zheng, Akiko Aizawa, Sadao Kurohashi

llms reinforcement-learning multimodal safety

arXiv AI

Sep 1

The Role of Network Topology and Opponent Information in Shaping Cooperation in Multi-Agent Reinforcement Learning Systems

arXiv:2608.28977v1 Announce Type: new Abstract: Several works have investigated the influence of graph topology on cooperation among artificial agents, while the majority of the literature has focuse...

By Seongho Son, Stephen Hailes, Mirco Musolesi

agents reinforcement-learning

arXiv Computer Vision

Sep 1

FuncRoom-Agent: Sequential Feed-Forward 3D Functional Indoor Scene Generation

arXiv:2608.29519v1 Announce Type: new Abstract: We introduce Function-Room Generation, a new indoor 3D scene generation setting that creates rooms supporting explicit functional goals rather than mer...

By Hao Feng, Zhi Zuo, MingJian Liang, Jingyu Hu, Xiaowei Hu, Liupengfei Wu, Dian Zhang, Guoxin Fang, Zhengzhe Liu

agents reinforcement-learning benchmarks

arXiv Machine Learning

Sep 1

Adaptive Doubly Robust Off-Policy Evaluation for Ranking Policies under Diverse User Behavior

The paper introduces Adaptive Doubly Robust (ADR), an off‑policy evaluation method for ranking policies that blends adaptive importance weighting with reward regression to reduce variance. ADR is unbiased when the true user behavior model is known and, under a sufficient condition, achieves lower variance than the prior Adaptive Inverse Propensity Scoring (AIPS) approach. Experiments on synthetic data show that ADR consistently improves mean squared error over AIPS and other ranking OPE estimators across various data sizes and ranking lengths.

By Kosuke Iguchi, Ren Kishimoto

reinforcement-learning safety

arXiv AI

Sep 1

Reference-Grafting Matches Fine-Tuning at Eliciting Sandbagged Capabilities

arXiv:2608.29458v1 Announce Type: cross Abstract: Sandbagging, in which a model deliberately underperforms on an evaluation despite retaining the underlying capability, threatens the safety evaluatio...

By Linh Le, Hong Kiat Tan, David Williams-King

llms reinforcement-learning fine-tuning safety

arXiv Computation and Language

Sep 1

Mind the Gap: Theory-of-Mind-Grounded Friction for Epistemic Alignment

arXiv:2608.30719v1 Announce Type: new Abstract: Productive dialogue alignment requires distinguishing \emph{surface coordination} (acknowledgments and smooth task progression) from \emph{epistemic al...

By Yifan Zhu, Kyeongmin Rim, James Pustejovsky

reinforcement-learning safety

arXiv AI

Sep 1

When Do Larger Batches Help Scale LLM Reinforcement Learning?

arXiv:2608.29296v1 Announce Type: cross Abstract: Larger batches reduce the variance of stochastic gradients per update and are therefore often expected to accelerate training. Yet whether this stati...

By Ziniu Li, Jinbo Wang, Guanhua Huang, Feiyuan Zhang, Pengbo Li, Alex Chen

llms reinforcement-learning efficiency

arXiv AI

Sep 1

Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence

arXiv:2608.31075v1 Announce Type: new Abstract: Recent advances in large reasoning models (LRMs) have shown that reinforcement learning with verifiable rewards (RLVR) can substantially improve reason...

By Zhiqin Yang, Jingwen Fu, Yuhan Liu, Hengyu Liu, Yonggang Zhang, Kainan Cao, Zizhuo Zhang, Chenxin Li, Ruibin Yuan, Jiahao Pan, Jiankai Sun, Zhenyuan Zhang, Yibo Li, Yunlong Lin, Jing Xiong, Sida Lin, Bo Han, Wei Xue, Yike Guo

agents reinforcement-learning

arXiv Machine Learning

Sep 1

What Emerges and What Breaks in Self-Play Driving

The paper reports on training autonomous driving policies via self‑play, extending previous work by using Transformers and a real‑city high‑definition map. On CARLA and Waymo benchmarks, the resulting policies underperform compared to Gigaflow, with identified failure modes such as reward hacking at traffic lights and lack of incentive to stop at stop signs. The authors also analyze which traffic rules emerge from self‑play and confirm that reward conditioning produces diverse driving behaviors.

By Laur Sisask, Ardi Tampuu, Tambet Matiisen

llms agents reinforcement-learning robotics benchmarks

Get the daily briefing