Multi-Task RL

RL: Reinforcement Learning

Multi-task reinforcement learning studies how an agent can learn several related objectives together, reusing experience to reduce data needs and transfer skills to new tasks. Common approaches include task-conditioned policies, shared representations and world models, meta-learning, successor features, and transformer or mixture-of-experts policies; curricula and distillation help allocate learning across tasks. Methods are tested in settings such as Meta-World manipulation, multi-agent benchmarks, robotics, network control, and multi-domain language-model training. A central challenge is balancing useful sharing against negative transfer: a model may improve on average while weakening on particular tasks or failing to generalize to unseen ones.

Latest papers

87 papers8 papers in the last four weeks, down 27% on the four weeks before. 0.0% of all new papers.

Collective Dynamics of Self-Reinforcement: How Cross-Task Coupling Drives Policy Collapse in Unsupervised RLVR
Oct 27, 2026Peize DingRL for Language Models Reinforcement Learning
Columbia University, New York, NY, USA
CONCUR: Concordant Auxiliary Rewards for Multi-Task, Multi-Reward Reinforcement Learning
Oct 27, 2026Jixuan Leng, Si Si, Fei Wang +4Reinforcement Learning RL for Language Models
CMU · Google DeepMind · Google +1
DoubleFree:Exploiting Exploration History for Rollout-Free Multi-Task LLMs Training
Oct 25, 2026Yingping Wang, Min Cai, Chenyu Yuan +3Multi-Teacher Knowledge Distillation RL Post-Training
The Hong Kong University of Science and Technology (Guangzhou) · University of Alberta
NetRL: Powering Network Decisions via NetGRPO-Driven Reinforcement Learning
Oct 12, 2026Lingfeng Peng, Xiaohui Xie, Yang Bao +1RL Control RL for LLM Agents
Tsinghua University
Teaching a Robot Dog New Tricks: Diverse Quadruped Skills via Combined Reinforcement and Imitation Learning with Adversarial Task Selection
Oct 6, 2026Lemon Foxmere, Anthony Furman, Yizheng Du +3Robot Skill Learning Multi-Task RL
HARE Lab, University of California at Santa Cruz, Santa Cruz, CA 95064 USA · AIEA Lab, University of California at Santa Cruz, Santa Cruz, CA 95064 USA
End-to-End Safe Social Navigation via Multi-Task Reinforcement Learning and Probabilistic Perception
Oct 5, 2026Tommaso Van Der Meer, Andrea Garulli, Antonio Giannitrapani +3Robotic RL Safe Robot Navigation
Dipartimento di Ingegneria dell’Informazione e Scienze Matematiche, Universit`a di Siena, Siena, Italy · Uninettuno University, Rome, Italy · Visual Intelligence for Transportation (VITA) Lab, ´Ecole Polytechnique F´ed´erale de Lausanne (EPFL), Lausanne, Switzerland
Task Inference Beyond Least Squares in Behavioral Foundation Models
Oct 4, 2026Kuan-Hsun Tu, Chien-Sheng Chiang, Hsin-Wei Chen +2Reinforcement Learning Multi-Task RL
National Taiwan University
RISED: RubrIcs for agentic multi-environment Selection and sElf-Distillation
Oct 1, 2026Jingtan Wang, Sirajul Salekin, Young mok Jung +3Reinforcement Learning Rubric-Based RL
National University of Singapore · Work done during an internship at Apple · Apple
Scalable Multi-Task Inverse Reinforcement Learning
Sep 30, 2026Allen Tran, Jia Wan, Nathan Kallus +1Reinforcement Learning Multi-Task RL
Netflix · Massachusetts Institute of Technology · Cornell Tech
Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RL
Sep 29, 2026Mathias Jackermeier, Jacques Cloete, Alessandro AbateRL Benchmarks Reinforcement Learning
Department of Computer Science, University of Oxford · Oxford Robotics Institute, University of Oxford
Uncertainty-Gated Exploration Noise Suppresses Task Collapse in Online RL Fine-Tuning of a Flow-Matching Vision-Language-Action Policy
Sep 23, 2026Mehmet Turan Yardımcı, Yunus Emre ÇoğurcuRobot Policy Learning RL Exploration
Karlsruhe Institute of Technology · Çukurova University
One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents
Sep 20, 2026Jie Zhao, Ziyu Jiang, Suhang Zheng +3Software Engineering Agents Agentic RL
Alibaba Group Model · Logics-SWE-Qwen3.6-27B
Stable and Efficient Real-World Online VLA Post-Training via Asynchronous Replay-Anchored Policy Improvement
Sep 19, 2026Jiarui Yang, Jiajin Zhang, Bin Zhu +2Robot Policy Learning RL Post-Training
Institute of Trustworthy Embodied AI, Fudan University · Shanghai Key Laboratory of Multimodal Embodied AI · Singapore Management University
UniExo: Unified Multi-Skill Policies for Musculoskeletal Locomotion and Co-Adaptive Exoskeleton Control
Sep 17, 2026Yifei Yuan, Jakob Wolf, Ghaith Androwis +1Robot Imitation Learning Assistive Robotics
Department of Biomedical Engineering, New Jersey Institute of Technology, Newark, NJ 07102, USA · Kessler Foundation, West Orange, NJ 07052, USA
Multitask Reinforcement Learning for Assisting Choice Model Specification
Sep 16, 2026Gabriel Nova, Stephane Hess, Sander Van CranenburghDiscrete Choice Modeling Deep Sets
CityAI Lab, Transport and Logistics Group, Delft University of Technology, The Netherlands · Institute for Transport Studies and Choice Modelling Centre, University of Leeds, UK
SkillX: Unified Multi-Skill Policy Learning for Humanoid Soccer
Sep 6, 2026Zhangchen Ye, Enxuan Ruan, Yifei Bao +10Robot Skill Learning Humanoid Robot Control
Noetix Robotics · Tsinghua University
One Step, One Lead: Mitigating Higher-Order Interference in Multi-Domain Reinforcement Learning via Cross-Step Control
Sep 6, 2026Zihan Lin, Xiaohan Wang, Jie Cao +4RL for Language Model Reasoning RL Post-Training
MAIS&NLPR, Institute of Automation, Chinese Academy of Sciences · Meituan, Beijing · School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences
Out-of-Distribution Generalisation with Sequence Models in Offline Multi-Agent Reinforcement Learning
Sep 3, 2026Oussama Hidaoui, Omer Ebead, Ulrich Armel Mbou Sob +14Zero-Shot Learning Reinforcement Learning
InstaDeep · AIMS · Stellenbosch University
WiSDoM: Wireless Sparse Decision Transformer with Mixture-of-Experts for Multi-Task Mobile Network Optimization
Aug 31, 2026Fatih Temiz, Shavbo Salehi, Melike Erol-KantarciSparse Mixture-of-Experts Decision Transformer
School of Electrical Engineering and Computer Science, University of Ottawa, Ottawa, Canada
T3S: Improving Multi-Task Reinforcement Learning with Task-Specific Feature Selector and Scheduler
Aug 31, 2026Yuanqiang Yu, Tianpei Yang, Yongliang Lv +2Robotic RL Multi-Task RL
College of Intelligence and Computing Tianjin University Tianjin, China · Department of Computing Science University of Alberta and Alberta Machine Intelligence Institute Edmonton, Canada
PAC: Progress-Augmented Advantage Curriculum for Multi-Task Reinforcement Learning of LLMs
Aug 31, 2026Yuanqiang Yu, Yanzhao Zheng, Zhentao Zhang +8RL for Language Model Reasoning Curriculum RL
Alibaba Group, Hangzhou, China
Consolidating RLVR Capabilities Across Domains: A Deep Dive into Fusion Paradigms
Aug 27, 2026Siye Wu, Kai Yang, Yuchen Cai +8Reinforcement Learning with Verifiable Rewards Language Model Merging
Fudan University · LLM Department, Tencent
Shared Actors Need Not Share Critics: Effects of Value Mismatch in Parallel Reinforcement Learning
Aug 27, 2026Zhenya Liu, Yang Meng, Zhuokai Zhao +2Reinforcement Learning Actor-Critic Methods
University of Chicago · University of Florida
Open-MOPD: Diagnosing and Fixing Capability Imbalance in Multi-Teacher On-Policy Distillation
Aug 19, 2026Huan-ang Gao, Haohan Chi, Yong Yan +7Multi-Teacher Knowledge Distillation On-Policy Distillation
SIA-Lab of Tsinghua AIR and ByteDance Seed · Institute for AI Industry Research (AIR), Tsinghua University · Department of Computer Science and Technology, Tsinghua University
Progressive Experience Fusion for Multi-Task World Model Control in Endovascular Navigation
Aug 19, 2026Harry Robertshaw, Maxence Boels, Nikola Fischer +4Robotic Endovascular Intervention Robot Policy Adaptation
Surgical & Interventional Engineering, School of Biomedical Engineering & Imaging Sciences, King’s College London, UK · Department of Neuroradiology, King’s College Hospital, UK
Task Specialization Fine-Tuning for Contextual Reinforcement Learning
Aug 17, 2026Jianan Zhou, Jung-Hoon Cho, Tianyue Zhou +5Contextual RL Reinforcement Learning
Nanyang Technological University · MIT · UIUC
Temporal Logic Guided Universal Task Representations for Reinforcement Learning
Aug 16, 2026Hao Zhang, Zhangli Zhou, Zhen KanRepresentation Learning for RL Reinforcement Learning
Department of Automation at the University of Science and Technology of China, Hefei, Anhui, China, 230026 · Anhui Provincial Key Laboratory of Humanoid Robots, Hefei, China, 230000
Control-Diverse Reinforcement Fine-Tuning: Decoupling the Shared Control Bottleneck of RL Post-Training
Aug 8, 2026Binwen Tan, Jingchao Wang, Dengzhe Hou +6RL for Language Model Reasoning Multi-Task RL
1Graduate School of Information Sciences, Tohoku University · School of Computer Science, Peking University · 2Unprecedented-scale Data Analytics Center, Tohoku University +3
SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs
Aug 4, 2026Kejian Zhu, Zhuoran Jin, Shangqing Tu +5Supervised Fine-Tuning Gradient Interference
The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, China · School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China · Tsinghua University, Beijing, China
HARGO: Heterogeneity-Aware Reward-Guided Optimization for RL Post-Training of LLMs on HPC Tasks
Jul 30, 2026Tiangang Li, Xiangbo TianRL for Language Models LLM Alignment
School of Computer Science, Wuhan University, Wuhan, China

Oct 27, 2026J/K move · Enter open · S save

Collective Dynamics of Self-Reinforcement: How Cross-Task Coupling Drives Policy Collapse in Unsupervised RLVR

Peize Ding

Columbia University, New York, NY, USA

RL for Language Models Reinforcement Learning RL Stability Unsupervised RL Reinforcement Learning with Verifiable Rewards Multi-Task RL

Open paper AskPDF arXiv

Label-free reinforcement learning with verifiable rewards (RLVR) rewards a language model for agreeing with its own majority vote. It improves reasoning at first but eventually collapses: the self-assessed reward goes to one while accuracy falls, often until the model gives the same answer to every problem. The collapse appears only when training uses many problems. We explain this dataset-size dependence as a collective instability. In a solvable many-problem model in which problems share parameters, a population whose problems prefer different answers amplifies any shared perturbation for as long as the population-averaged slope

\langle F'\rangle

of the single-problem self-reinforcement drift

F

is positive. This holds while problems are still uncertain about their answers. Per-problem sharpening, slowed by

1/N

for

N

training problems, races against the shared mode, which grows at a rate

\proptoλ^2

independent of

N

, where

λ

is the strength of the shared coupling. This yields a critical number of problems

N_c\simeq\ln(1/δ)/(λ^2G)

, where

δ

seeds the shared mode and

G

decreases with the initial confidence. Simulations of the model confirm the predictions: a sharp transition,

N_c\proptoλ^{-2}

up to a logarithmic correction, a single critical amplification

λ^2NG

across couplings, confidence profiles and training variants, a logarithmic dependence on batch size, and no collapse for confident initial models. A kernel version of the theory identifies a protective coupling (generalization of sharpening) and a marginal-entropy bonus that stabilizes the uniform collective mode. In label-free RL of a small Transformer, removing the shared direction prevents collapse entirely, restoring it causes collapse, and collapse grows with training-set size and base-model entropy.