Reinforcement learning

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

6,856 stories · RSS feed

arXiv AI

Sep 15

VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding

arXiv:2609.15606v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable progress on short video understanding yet remain limited on long videos due to the...

By Weixin Xu, Zhenyu Yang, Bing Wang, Shengsheng Qian, Changsheng Xu

llms agents reinforcement-learning fine-tuning multimodal benchmarks safety

arXiv AI

Sep 15

Learning Human-Like Badminton Skills for Humanoid Robots

arXiv:2602.08370v2 Announce Type: replace-cross Abstract: Realizing versatile and human-like performance in high-demand sports like badminton remains a formidable challenge for humanoid robotics. Unl...

By Yeke Chen, Shihao Dong, Xiaoyu Ji, Jingkai Sun, Zeren Luo, Liu Zhao, Jiahui Zhang, Wanyue Li, Ji Ma, Bowen Xu, Yimin Han, Xuanyi Li, Yudong Zhao, Liyun Li, Peng Lu

reinforcement-learning robotics efficiency safety

arXiv Computation and Language

Sep 15

HumorGen: Cognitive Synergy for Humor Generation in Large Language Models via Persona-Based Distillation

arXiv:2604.09629v3 Announce Type: replace Abstract: Humor generation poses a significant challenge for Large Language Models (LLMs), because their standard training objective (next-token prediction)...

By Edward Ajayi, Prasenjit Mitra

llms reinforcement-learning fine-tuning efficiency safety

arXiv Machine Learning

Sep 15

Hybrid Feedback-Guided Optimal Learning for Wireless Interactive Panoramic Scene Delivery

arXiv:2602.07273v2 Announce Type: replace Abstract: Immersive applications such as virtual and augmented reality impose stringent requirements on frame rate, latency, and synchronization between phys...

By Xiaoyi Wu, Juaren Steiger, Bin Li, R. Srikant

reinforcement-learning benchmarks

arXiv Computation and Language

Sep 15

All Learning Has an Emotional Basis, So Does Task-Oriented Dialogue

arXiv:2507.01594v2 Announce Type: replace Abstract: Task-oriented dialogue (ToD) systems aim to help users accomplish goals through natural language interaction. Beyond task success, effective ToD sy...

By Shutong Feng, Hsien-chin Lin, Nurul Lubis, Carel van Niekerk, Michael Heck, Benjamin Ruppik, Renato Vukovic, Milica Ga\v{s}i\'c

llms reinforcement-learning

arXiv AI

Sep 15

ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search

ZGCM-1 is a 7B dense foundation model trained from scratch with extreme data, system, and algorithmic efficiency. It uses a core premise that compact models can overcome capacity limits by combining deliberate internal thinking with active external tool use, supported by a 256K context and an end‑to‑end high‑efficiency training recipe that includes interleaved gated sliding‑window and full attention, a stable FP8 Muon optimizer, progressive curriculum scaling, and reformulation of interaction traces into Markov Decision Processes. The model is competitive with much larger frontier models on challenging mathematical reasoning and agentic search tasks, offers a ~4.2× efficiency improvement in pre‑training time‑to‑loss, and its weights, checkpoints, training code, data recipes, and logs are fully open‑source to support community research.

By Jiyan He, Guang Liang, Hao Liu, Haoxiang Guan, Jinbo Sun, Junyi Guo, Wenjun Feng, Yantai Xie, Yifei Shen, Bin Shao, Chuyang Wei, Kai Chen, Kexin Zhou, Minghang Zhu, Shuxin Zheng, Tie-Yan Liu, Taine Zhao, Wenhui Zhu, Xueyin Xu, Xiaoqing Zhang, Yatao Li, Yuxuan Ren

agents reinforcement-learning fine-tuning efficiency benchmarks

arXiv AI

Sep 15

Privileged observations enable rapid and reliable policy discovery directly in the physical world

arXiv:2512.08463v2 Announce Type: replace Abstract: We study how privileged information about a physical system affects the discovery of high-performing policies when training a reinforcement learnin...

By Antonio Terpin, Raffaello D'Andrea

agents reinforcement-learning

arXiv AI

Sep 15

Shallow Beliefs: Synthetic document finetuning does not inoculate against emergent misalignment from reward hacking

arXiv:2609.14998v1 Announce Type: new Abstract: Recent work shows that models that learn to reward hack on RL environments can become broadly misaligned, and that reframing reward hacking as acceptab...

By Arun Jose, Julian Stastny

llms reinforcement-learning fine-tuning

arXiv AI

Sep 15

TryOnReward: Learning Foveated Consistency for Reinforcement Fine-Tuning of Virtual Try-On

arXiv:2609.13259v1 Announce Type: cross Abstract: Virtual Try-On (VTON) aims to dress a person with the reference garment, producing visually reasonable results aligned with human preferences. Turnin...

By Xueheng Li, Yong Liu, Xiaolong Fu, Wen Xue, Chengjun Xie, Yipeng Sun, Yan Li, Simiu Gu

reinforcement-learning fine-tuning multimodal benchmarks safety

arXiv Computation and Language

Sep 15

Inoculation Midtraining with Learned Neologisms

arXiv:2609.15886v1 Announce Type: new Abstract: Large language models (LLMs) often learn both desirable and undesirable properties during post-training. We study whether midtraining, an earlier train...

By Kyle O'Brien, Edward James Young, Puria Radmard, Nathalie Kirch, Cameron Tice, Tomek Korbak, David Demitri Africa

llms reinforcement-learning fine-tuning safety

arXiv AI

Sep 15

Not All Prompts Are Equal: Exploration-Guided Prompt Scaffolding for Multimodal Reinforcement Post-Training

The paper introduces an exploration-guided prompt scaffolding framework for multimodal large language models, dynamically adjusting the prompt distribution during reinforcement learning post-training. It uses an Exploration Potential Score (EPS) derived from KL-regularized policy improvement to assess prompt utility without extra overhead, and a teacher model rewrites low-utility prompts to preserve intent while improving informativeness. Experiments on Geo3K, MMK12, MathVision, and MMMU-Pro show consistent performance gains, up to 9.7% in-domain and over 11% on out-of-distribution benchmarks.

By Yuanhao Yue, Qianli Ma, Chengyu Wang, Haoting Wang, Lei Shen, Jun Huang

llms reinforcement-learning multimodal benchmarks

arXiv Machine Learning

Sep 15

High-Probability Nash Regret for Decentralized Learning in Markov $\alpha$-Potential Games: Episodic and Fully Online Asynchronous Algorithms with Applications to Markov Congestion Games

arXiv:2609. 14959v1 Announce Type: new Abstract: We study decentralized learning of Nash equilibria (NE) in infinite-horizon discounted Markov games under bandit feedback, focusing on Markov $\alpha$-potential games.

By S. Rasoul Etesami

reinforcement-learning safety

arXiv AI

Sep 15

Diagnosing Faults in Reinforcement Learning Simulators and World Models with Canonical Polynomial Invariants

arXiv:2609.13194v1 Announce Type: cross Abstract: A large literature builds physical structure into learned dynamics on the premise that models respecting the underlying physics predict better. We te...

By Tesfay Zemuy Gebrekidan, Hadush Hailu Gebrerufael

reinforcement-learning benchmarks

arXiv Computer Vision

Sep 15

Long-to-Short Video Evidence Reasoning for Grounded Question Answering

arXiv:2609.15224v1 Announce Type: new Abstract: We present LOVER, a \underline{L}ong to sh\underline{O}rt \underline{V}ideo \underline{E}vidence \underline{R}einforced model for grounded question ans...

By Kaiyan Chen, Junbin Xiao, Xun Yang

nlp reinforcement-learning benchmarks

arXiv Machine Learning

Sep 15

Learning Steerable Clarification Policies with Collaborative Self-play

arXiv:2512.04068v3 Announce Type: replace Abstract: To handle underspecified or ambiguous queries, AI assistants need a policy for managing their uncertainty to determine (a) when to guess the user i...

By Jonathan Berant, Maximillian Chen, Adam Fisch, Reza Aghajani, Fantine Huot, Mirella Lapata, Jacob Eisenstein

agents reinforcement-learning

arXiv Machine Learning

Sep 15

Prelimit Coupling and Steady-State Convergence of Constant-stepsize Nonsmooth Contractive SA

arXiv:2404.06023v3 Announce Type: replace-cross Abstract: Motivated by Q-learning, we study nonsmooth contractive stochastic approximation (SA) with constant stepsize. We focus on two important class...

By Yixuan Zhang, Dongyan Huo, Yudong Chen, Qiaomin Xie

reinforcement-learning safety

arXiv Machine Learning

Sep 15

From Linear to Linearizable Optimization: A Novel Framework with Applications to Stationary and Non-stationary DR-submodular Optimization

arXiv:2405.00065v4 Announce Type: replace-cross Abstract: This paper introduces the notion of upper-linearizable/quadratizable functions, a class that extends concavity and DR-submodularity in variou...

By Mohammad Pedramfar, Vaneet Aggarwal

reinforcement-learning benchmarks

arXiv Machine Learning

Sep 15

Steady-State Convergence of Stochastic Approximation

arXiv:2609.14922v1 Announce Type: cross Abstract: For constant-stepsize stochastic approximation (SA), the iterates converge in distribution to a stationary law that depends on the stepsize $\alpha.$...

By Yixuan Zhang, Qiaomin Xie

reinforcement-learning safety

arXiv AI

Sep 15

Planning or Learning: Reliability and Cost in Multi-Asset Maintenance

arXiv:2609.13566v1 Announce Type: new Abstract: Industrial maintenance systems involve multiple interacting assets and shared resources, making it challenging to balance reliability and operational c...

By Xian Yeow Lee, Chandrasekar Venkatraman, Ahmed Farahat

agents reinforcement-learning benchmarks

arXiv Machine Learning

Sep 15

Strong and Compact Policies for Submodular Markov Decision Processes via LP-Based Submodular Orienteering

arXiv:2609.15539v1 Announce Type: cross Abstract: Finding policies for Markov Decision Processes (MDPs) is a central problem in areas such as Reinforcement Learning and Operations Research. Here, we...

By Lars Rohwedder, Rico Zenklusen

agents reinforcement-learning

Get the daily briefing