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
Position: CNNs Don't See Shape — And That Won't Change Without New Architectures
Position: Code Benchmarks Should Prioritize Rigor, Reliability, and Reproducibility
Position: Evaluating LLMs in Finance Requires Explicit Bias Consideration
Position: If Open Source Is to Win, It Must Go Public
Position Is All You Need: A Free Lunch Token Compression Strategy for MLLM-based Referring Expression Segmentation
Position: Machine Learning for Heart Transplant Allocation Policy Optimization Should Account for Incentives
Position: Medical AI Neglects Real Treatment Outcomes
Position: Preventing AI-Generated CSAM Necessitates New Approaches to AI Safety
Position: Reasoning After Perception Means Reasoning Without Vision
Position: RLHF May Not Reflect Genuine Preferences
Position: RL Researchers Need to Distinguish Between Solving Simulators and Using Simulators as a Proxy
Position: Stop Chasing the C-index when Evaluating Survival Analysis Models
Position: Stop Preaching and Start Practising Data Frugality for Responsible Development of AI
Position: Theory-Level Autoformalization, From Isolated Statements to Unified Formal Knowledge Bases
Position: There are futures that benchmark-driven AI cannot see
Position: The Systemic Lack of Agency in Visual Reasoning
Position: Time-Series Foundation Models Require Explicit Domain-Level Benchmarks
Position: Uncertainty is a Strategic Signal in Human–AI Decision Making
Position: Virtual Cells Need Context, Not Just Scale
Position: What Cézanne Knew About Visual Intelligence That Vision-Language Models Miss
Position: When AI Decides Who Gets an Organ: Multi-Agentic AI Systems in Transplant Medicine Risk Amplifying Disparities Without Targeted Explainability and Deployment Strategies
Positive Distribution Shift as a Framework for Understanding Tractable Learning
Positive-Unlabeled Learning with Extreme Scarcity of Labeled Positives
Positive–Unlabeled Reinforcement Learning Distillation for On-Premise Small Models
Possibilistic Predictive Uncertainty for Deep Learning
PosterAgent: Agentic Poster Generation via Stage-Aware Reinforcement Learning
Posterior Behavioral Cloning: Pretraining BC Policies for Efficient RL Finetuning
Posterior Concentration of Bayesian Physics-Informed Neural Networks for Elliptic PDEs
Posterior Mismatch Matters: Adversarial Training for Long-Tailed Robustness
Posterior Sampling Reinforcement Learning with Gaussian Processes for Continuous Control: Sublinear Regret Bounds for Unbounded State Spaces
Post-Hoc Merging is Not Enough: Many-Shot Model Merging with Loss-Gap Balancing
PostTrainBench: Can LLM Agents Automate LLM Post-Training?
Post-Training Language Models for Crosslingual Consistency
Post-Training LLMs as Better Decision-Making Agents: A Regret-Minimization Approach
Post-Training with Policy Gradients: Optimality and the Base Model Barrier
Power-Boosted Granger-Causal Discovery for Large Heterogeneous Panel Data
PowerFlow: Unlocking the Dual Nature of LLMs via Principled Distribution Matching
Powerful and Theoretically Guaranteed Independence Testing on Heterogeneous Federated Clients
PPDL: LLM-Based Flows as Probabilistic Programs
PPI Candidate Ranking: Large-Scale Evaluation of a Domain Knowledge–Guided Pipeline
PPT-Eval: A Benchmark for Computer-Use Agents on PowerPoint Tasks
PRAC: Principal-Random Subspace for LLM Activation Compression and Memory-Efficient Training
Practical and Optimal Algorithm for Linear Contextual Bandits with Rare Parameter Updates
Practical and Scalable Hamiltonian Monte Carlo Without the Metropolis Test
Practical Mechanism for Fault-Tolerant Spiking Neural Networks via Simple Input Control Based on Learnable Fragmentation
PragLocker: Protecting Agent Intellectual Property in Untrusted Deployments via Non-Portable Prompts
Precision-Induced Miscalibration: Understanding and Correcting Confidence Distortion in Quantized Neural Networks
Preconditioned DeltaNet: Curvature-aware Sequence Modeling for Linear Recurrences
Preconditioning Neural Tangent Kernel for Adaptive Optimization
Predictable Compression Failures: Order Sensitivity and Information Budgeting for Evidence-Grounded Binary Adjudication

« prev Page 86/130 next »