Compositional Generalization

Follow this topic Ask about this topic

Momentum

6 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

Jul 6Week of Sep 21

Latest papers 46

All topics

On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D
Sep 30, 2026Fabio J. Fehr, Philip TorrHypernetwork Model Weights
University of Oxford
Same Scene, Different Task: Skill Alignment for Compositional Generalization in VLAs
Sep 30, 2026Taegeun Yang, Youngju Na, Yoonki Cho +1Photometric Supervision Compositional Generalization
Korea Advanced Institute of Science and Technology (KAIST)
Correcting WHERE, Preserving HOW: Compositional Generalization for Vision-Language-Action Models via Referential Guidance
Sep 29, 2026Yanyan Zhang, Disheng Liu, Xinpeng Li +8Large-Scale Robot Demonstration Datasets Action Generation
Case Western Reserve University Cleveland, OH, USA
A Spectral Theory of Compositional Learning
Sep 27, 2026Hugo RydelCompositional Generalization Cognitive Science
University of Manchester
Compositional Reasoning in Language Models under Reinforcement Learning Post-Training
Sep 16, 2026Yu He, Yingxi Li, Yifei Wang +1Compositional Generalization Reinforcement Learning Post-Training
Stanford University · Amazon AGI Labs
What You Can't See Is Still What You Learn: A Preregistered Sixty-Society Confirmation That Evidence Masking Drives Compositional Generalization
Sep 15, 2026Narcis MarincatCompositional Generalization Ablation
Independent researcher.
Neuro-Symbolic Hierarchical Intention Anticipation in Human Behavior
Sep 15, 2026Farnaz Soleimani, Abdelghani Chibani, Yacine Amirat +1Neuro-Symbolic Framework Agent Behavior Modeling
LISSI · aLISSI Laboratory, University of Paris-Est Créteil (UPEC), IUT de Créteil-Vitry, France
Type Diversity Enables Transformers to Generalise Compositionally
Sep 14, 2026Anssi Moisio, Mathias Creutz, Mikko KurimoCompositional Generalization Diversity
Aalto University · University of Helsinki
CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization
Sep 9, 2026Luyao Tang, Jiewei Zheng, Kunze Huang +3Generalized Category Discovery Compositional Generalization
The University of Hong Kong, Hong Kong SAR, China · Xiamen University, Xiamen, China · Shenzhen University, Shenzhen, China.
Compositional Failure in Audio-Visual LLMs: Late-Layer Prior Dominance Under Cross-modal Conflict
Aug 27, 2026Adarsh Sudheer, David Li, Omar Elbanna +3Cross-Modal Conflicts Large Audio Language Models
Independent Researcher.
Learning Generalizable Behaviors for Terminal Agents
Aug 23, 2026Yihang Yao, Bo Pang, Xuan Phi Nguyen +3Agentic Reinforcement Learning Agentic Benchmarks
Salesforce AI Research · Carnegie Mellon University
Explicit Language Memory for Long-Horizon Planning in Vision-Language-Action Models
Aug 5, 2026Houze Xu, Jizhong Li, Ziyi YeDiffusion-Based Vision-Language-Actions Long-Horizon Planning
Fudan University
Diagnosing Compositional Generalization in Sequential Robot Tasks
Jul 31, 2026Yixiao Wang, Cheng-En Wu, Lingfeng Sun +5Compositional Generalization Robotic Manipulation
University of California, Berkeley · Tsinghua University
Kimi K3: Open Frontier Intelligence
Jul 27, 2026Kimi Team, Tongtong Bai, Yifan Bai +399Qwen3 Intelligence
Scale Up Strategically: Learning Compositional Generalization via Bias-Aware Evaluation and Data Collection for Robotic Manipulation
Jul 23, 2026Yu Qi, Zhang Ye, Xinyi Xu +6Learning Policy Robotic Manipulation
Northeastern University · NVIDIA
On the Computational Complexity of Structural Generalization
Jul 21, 2026Zichao WeiCompositional Generalization Learnability
Department of Language Science and Technology Saarland University Saarbrücken, Germany
Progressive Reasoning with Primitive Correction for Compositional Zero-Shot Learning
Jul 7, 2026Ziyi Chen, Haoyan Shi, Sunhan Xu +1Compositional Generalization Progressive Reasoning
School of Computer Science and Technology, Beijing Jiaotong University · Key Laboratory of Big Data & Artificial Intelligence in Transportation (Ministry of Education)
ClassicLogic: A Knowledge-Driven Benchmark of Classic Puzzle Games for Evaluating Compositional Generalization
Jul 6, 2026Mahnoor Shahid, Hannes RothePuzzle Reasoning Benchmark
Universität Duisburg-Essen Germany
InternVLA-A1.5: Unifying Understanding, Latent Foresight, and Action for Compositional Generalization
Jul 6, 2026Haoxiang Ma, Junhao Cai, Xiaoxu Xu +26Action Generation Foresight
Physical Intelligence Team, Shanghai AI Laboratory
Open-Set Source Tracing as Compositional Factors via Structured Prototypes
Jul 3, 2026Santiago Rubio, Antonio Almudévar, Antonio Miguel +2Speech Separation Open-Set Recognition
ViVoLab, I3A University of Zaragoza Zaragoza, Spain
Learning to Reason with Curriculum II: Compositional Generalization
Jun 26, 2026Nived Rajaraman, Audrey Huang, Miroslav Dudik +3Adaptive Curriculum Compositional Generalization
Microsoft Research. · University of Illinois Urbana-Champaign. · Microsoft Research, New York, NY.
From Reasoning Traces to Reusable Modules: Understanding Compositional Generalization in Language Model Reasoning
Jun 16, 2026Lingjing Kong, Xin Liu, Guangyi Chen +9LLM Reasoning Strategies Compositional Generalization
Carnegie Mellon University · Mohamed bin Zayed University of Artificial Intelligence · Institute of Foundation Models +1
ATOM-Bench: A Real-World Benchmark for Atomic Skills and Compositional Generalization in Manipulation Policies
Jun 15, 2026Zenan Wu, Bingqing Wei, Lu Liu +8Robotic Manipulation Policies Compositional Generalization
1Beijing Academy of Artificial Intelligence · 2Peking University
A Compositional Framework for Open-ended Intelligence
Jun 13, 2026Ida Momennejad, Roberta RaileanuCompositional Generalization General Intelligence
Microsoft Research NYC, New York, USA · Google DeepMind, London, UK.
Assign and Add: A Mechanistic Study of Compositional Arithmetic
May 29, 2026Brady Exoo, Alberto Bietti, John SousCompositional Generalization Transformer Architectures
Yale University · Flatiron Institute
Compositional Generalization in Autoregressive Models via Logit Composition
May 27, 2026Aakash Kumar, Maria Sofia Bucarelli, Emanuele NataleCompositional Generalization Autoregressive Model
COATI, CNRS, Inria, I3S, Université Coté d’Azur, France
Compositional Transduction with Latent Analogies for Offline Goal-Conditioned Reinforcement Learning
May 20, 2026Junseok Kim, Dohyeong Kim, Mineui Hong +1Goal-Conditioned Reinforcement Learning Compositional Generalization
Department of Electrical and Computer Engineering and ASRI, Seoul National University · Independent researcher · Robotics Institute, Carnegie Mellon University
Matrix-Space Reinforcement Learning for Reusing Local Transition Geometry
May 14, 2026Zuyuan Zhang, Carlee Joe-Wong, Tian LanOffline Reinforcement Learning Transitions
The George Washington University · Carnegie Mellon University
Unlocking Compositional Generalization in Continual Few-Shot Learning
May 12, 2026Phu-Quy Nguyen-Lam, Phu-Hoa Pham, Dao Sy Duy Minh +3Few-Shot Learning Self-Supervised Vision Transformers
Faculty of Information Technology, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam. · Department of Computer Science, University of Warwick, Coventry, United Kingdom.
Benchmarking Compositional Generalisation for Machine Learning Interatomic Potentials
May 9, 2026Amir Masoud Nourollah, Irtaza Khalid, Stefano Leoni +1Interatomic Potentials Molecular Property Prediction
Cardiff University Cardiff, CF10 3AT, UK
Test-Time Compositional Generalization in Diffusion Models via Concept Discovery
May 8, 2026Zekun Wang, Anant Gupta, Tianyi Zhu +1Diffusion Models Compositional Generalization
Georgia Institute of Technology · University of Virginia
Reinforcement Learning for Compositional Generalization with Outcome-Level Optimization
May 6, 2026Xiyan Fu, Wei LiuCompositional Generalization Feedback
Nanyang Technological University · Independent Researcher
Critical Windows of Complexity Control: When Transformers Decide to Reason or Memorize
May 6, 2026Sarwan AliTransformer Architectures Weight Decay
Columbia University, Irving Medical Center, NY, USA
Investigating More Explainable and Partition-Free Compositionality Estimation for LLMs: A Rule-Generation Perspective
Apr 30, 2026Ziyao Xu, Cong Wang, Houfeng WangCompositional Generalization Compositionality
National Key Laboratory for Multimedia Information Processing, School of Computer Science, Peking University · OPPO AI Center
Regression Test Selection for Updated Capability Modules in Compositional ML Systems via Atomic-Quality Probes
Apr 29, 2026Xue Qin, Simin Luan, Cong Yang +1Compositional Generalization Text Dominance
1*School of Software, Harbin Institute of Technology, Harbin, China. · School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China. · School of Future Science and Engineering, Soochow University,2026 Suzhou, China.
AGEL-Comp: A Neuro-Symbolic Framework for Compositional Generalization in Interactive Agents
Apr 29, 2026Mahnoor Shahid, Hannes RotheNeuro-Symbolic Framework Compositional Generalization
Universität Duisburg-Essen, Essen, Germany
Grounding vs. Compositionality: On the Non-Complementarity of Reasoning in Neuro-Symbolic Systems
Apr 29, 2026Mahnoor Shahid, Hannes RotheNeuro-Symbolic Framework Compositional Generalization
Universität Duisburg Essen, Germany
Gradient-Based Program Synthesis with Neurally Interpreted Languages
Apr 20, 2026Matthew V. Macfarlane, Clément Bonnet, Herke van Hoof +1Llm-Driven Code Synthesis Neuro-Symbolic Framework
AMLab, University of Amsterdam · Work completed while a visiting student at the University of Alberta. · Ndea +2
$π_{0.7}$ : a Steerable Generalist Robotic Foundation Model with Emergent Capabilities
Apr 16, 2026Physical Intelligence, Bo Ai, Ali Amin +85Scalable Robot Learning Cross-Embodiment Transfer
Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models
Feb 27, 2026Arnas Uselis, Andrea Dittadi, Seong Joon OhCompositional Generalization Neural Representations
Tübingen AI Center, University of Tübingen · Helmholtz · Technical University of Munich
Why Can't I Open My Drawer? Mitigating Object-Driven Shortcuts in Zero-Shot Compositional Action Recognition
Jan 22, 2026Geo Ahn, Inwoong Lee, Taeoh Kim +3Open-Vocabulary Action Recognition Compositional Generalization
1Kyung Hee University · ∗This work was done during an internship at NAVER Cloud. · 2NAVER Cloud
Mixtures of SubExperts for Large Language Continual Learning
Nov 9, 2025Haeyong Kang, Hee Suk Yoon, Dahua Feng +1Replay-Based Continual Learning Parameter-Efficient Fine-Tuning Methods
Korea Advanced Institute of Science and Technology (KAIST)
What Drives Compositional Generalization in Visual Generative Models? The Importance of Continuous Training Objectives
Oct 3, 2025Karim Farid, Rajat Sahay, Yumna Ali Alnaggar +4Compositional Generalization Generative Models
University of Freiburg · Bosch Center for Artificial Intelligence · Inria, École Normale Supérieure, CNRS, PSL Research University
Learning Composable Chains-of-Thought
May 28, 2025Fangcong Yin, Zeyu Leo Liu, Liu Leqi +2LLM Reasoning Strategies Chain-of-Thought Reasoning
The University of Texas at Austin · Princeton University
Disassociating performance from compositional feature learning
May 14, 2025George Dimitriadis, Spyridon SamothrakisCompositional Generalization Out-Of-Distribution
Computer Science and Electronic Engineering Department of the University of Essex
A Theoretical Analysis of Provable Compositional Generalization in Neural Networks: A Necessary and Sufficient Condition
May 5, 2025Yuanpeng LiCompositional Generalization Neural Network

Sep 30, 2026, cs.LGJ/K move · Enter open · S save

On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D

Fabio J. Fehr, Philip Torr

University of Oxford

Hypernetwork Model Weights Few-Shot Learning Generative Architectures Improved Generalization Compositional Generalization+1

Open paper AskPDF arXiv

General-purpose models can adapt to many tasks from context, while specialised models can execute individual functions with less capacity. Yet obtaining such specialists requires task-specific training or adaptation. We ask whether they can instead be generated directly from a few demonstrations. Using ARC-1D as a controlled testbed, we show that individual transformations can be represented by tiny specialist models, and that a hypernetwork can generate their parameters from context. The generated parameters form a structured weight space, while the resulting specialists show partial compositional generalisation and generalisation to transformations not seen during training. In both settings, removing explicit task identifiers improves generalisation beyond the training transformations. Together, these results provide a proof of concept that few-shot task context can be compiled on-the-fly into compact executable model parameters, and that the resulting weight space can support reuse and generalisation beyond known functions.

Momentum

Related topics

Latest papers 46