ICML 2026

6,487 papers

Top Research Directions

Research directions most represented by ICML 2026 papers.

Scaling Language Model Reasoning Beyond Verifiable Rewards 79
Group Relative Policy Optimization for Large Reasoning Models 63
General Purpose Reasoning from Specialized Reinforcement Learning 62
Overcoming the Iterative Inference Latency of Diffusion Models 50
Shifting from Static Benchmarks to Dynamic Evaluation 47
Controlling Language Models Through Latent Space Interventions 45
Resolving Superposition with Sparse Autoencoders 42
Autonomous Code Refinement via Execution Feedback Loops 41
The Performance Cliff in Ultra Low Bit Quantization 41
Activation Steering for Inference Time Model Control 39
Mitigating Expert Collapse in Sparse Mixture of Experts 37
Proxy Objective Exploitation in Language Model Alignment 37
The Decoupling of Visual Perception and Logical Reasoning 36
Mitigating Computational Overthinking in Large Reasoning Models 35
Input Adaptive Compute Allocation for Language Models 33

Top Authors

AuthorPapers
Junchi Yan19
Min Zhang19
Tat-Seng Chua15
Haoxuan Li14
Philip Torr14
Yong Li14
Kun Zhang13
Li Shen13
Tianlong Chen13
Xiang Wang13
Xin Geng13
Yejin Choi13
Bo An12
Mihaela van der Schaar12
Xunliang Cai12
Bo Han11
Ge Zhang11
Han Liu11
Jiang Bian11
LEI BAI11

Impactful Papers

Papers from ICML 2026 that shaped research clusters.

Foundational

Papers

Title
Learning to Evict from Key-Value Cache
Learning to Execute Graph Algorithms Exactly with Graph Neural Networks
Learning to Explore: Scaling Agentic Reasoning via Exploration-Aware Policy Optimization
Learning to Extrapolate to New Tasks: A Relational Approach to Task Extrapolation
Learning to Label: A Reinforced Self-Evolving Framework for Semi-supervised Referring Expression Segmentation
Learning-To-Measure: In-Context Active Feature Acquisition
Learning to Memorize with Attributive and Associative Memory for Online Test-Time Adaptation of Vision-Language Models
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs
Learning-to-Optimize via Deep Unfolded Flows
Learning to Perceive the World Through Control: Empowerment-Based Representation Learning
Learning to Rank by Directly Optimizing Full-Order Probabilities
Learning to Rank from Incomplete Rankings
Learning to Reason for Factuality
Learning to Reconfigure: Configuration-Control Co-optimization of Reconfigurable Robots for Heterogeneous Locomotion
Learning to Refine: Spectral-Decoupled Iterative Refinement Framework for Precipitation Nowcasting
Learning to Remember, Learn, and Forget in Attention-Based Models
Learning to Route Languages for Multilingual Policy Optimization
Learning to Search and Searching to Learn for Generalization in Planning
Learning to Self-Verify Makes Language Models Better Reasoners
Learning to Share: Selective Memory for Efficient Parallel Agentic Systems
Learning to Theorize the World from Observation
Learning to Think in Physics: Breaking Shortcut Learning in Scientific Diffusion via Representation Alignment
Learning to Watch: Active Video Anomaly Understanding via Interleaved Policy Optimization
Learning to Watermark in the Latent Space of Generative Models
Learning Transferable Interaction Primitives from Game Videos for Humanoid Locomotion
Learning Treatment Allocations with Risk Control Under Partial Identifiability
Learning Treatment Representations for Downstream Instrumental Variable Regression
Learning Unanimously Acceptable Lotteries via Queries
Learning Unmasking Policies for Diffusion Language Models
Learning What to Generate: A Reinforcement Learning-based Closed-Loop Augmentation Framework for Person Re-identification
Learning When to Act or Refuse: Guarding Agentic Reasoning Models for Safe Multi-Step Tool Use
Learning When to Attend: Conditional Memory Access for Long-Context LLMs
Learning with Admissibility: Robust Fuzzy Hashing for Cross-Modal Retrieval with Noisy Labels
Learn to change the world: Multi-level reinforcement learning with model-changing actions
Learn-to-learn on Arbitrary Textual Conditioning: A Hypernetwork-Driven Meta-gated LLM
Learn to Merge: Meta-Learning for Adaptive Multi-Task Model Merging
Learn to Think: Improving Multimodal Reasoning through Vision-Aware Self-Improvement Training
Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts
LEC: Linear Expectation Constraints for Selection-Conditioned Risk Control in Selective Prediction and Routing Systems
LECTOR: Joint Learning of Scientific Reasoning Graphs and Introduction Generation
LECTOR: Joint Optimization of Scientific Reasoning Graphs and Introduction Generation
Left–Right Symmetry Breaking in CLIP-Style Vision-Language Models Trained on Synthetic Spatial-Relation Data
LEGO: An LLM-Enabled Hierarchical Optimizer for Tensor Computation Graphs with Structure-Aware Search and Compositional Synthesis
LEGO-FL: Learning Heterogeneous Federated Models as a LEGO Assembly Games
LEMUR: Learned Multi-Vector Retrieval
Length Generalization Bounds for Transformers
LERD: Latent Event-Relational Dynamics for Neurodegenerative Classification
Less Data, Faster Training: repeating smaller datasets speeds up learning via sampling biases
Less Diverse, Less Safe: The Indirect But Pervasive Risk of Test-Time Scaling in Large Language Models
Less is Enough: Synthesizing Diverse Data in LLM Feature Space with Sparse Autoencoders

« prev Page 64/130 next »