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
When to Memorize and When to Stop: Gated Recurrent Memory for Long-Context Reasoning
When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning
When to Trust the Cheap Check: Weak and Strong Verification for Reasoning
Where Concept Erasure Should Occur: Concept–Layer Alignment in Text-to-Video Diffusion Models
Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection
Where Flow Matching Leaks: Characterising the Membership Signals Along the Interpolation Path
Where Rectified Flows Leak: Characterising Membership Signals Along the Interpolation Path
Which Algorithms Can Graph Neural Networks Learn?
Which Heads Matter for Reasoning? RL-Guided KV Cache Compression
WhisperSplat: Lossless Steganography in 3D Gaussian Splatting
Who can we trust? LLM-as-a-jury for Comparative Assessment
Who Evaluates AI's Social Impacts? Mapping Coverage and Gaps in First and Third Party Evaluations
Who Gets Credit or Blame? Attributing Accountability in Modern AI Systems
Whom to Query for What: Adaptive Group Elicitation via Multi-Turn LLM Interactions
Who Said Neural Networks Aren't Linear?
Who’s in Charge? Disempowerment Patterns in Real-World LLM Usage
Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons
Why Agentic Theorem Prover Works: A Statistical Provability Theory of Mathematical Reasoning Models
Why Are Linear RNNs More Parallelizable?
Why DDIM Hallucinates More Than DDPM: A Theoretical Analysis of Reverse Dynamics
Why Dedicated Critics: Eliminating Target Drift in Multi-Constraint RL
Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment
Why Do We Need Warm-up? A Theoretical Perspective
Why Linear Recurrent Memory Works in Partially Observable Reinforcement Learning
Why ReLU? A Bit-Model Dichotomy for Deep Network Training
Why Self-Distillation Helps and Hurts: Denoising vs. Signal Forgetting
Why Specialist Models Still Matter: A Heterogeneous Multi-Agent Paradigm for Medical Artificial Intelligence
Why Tree-Style Branching Matters for Thought Advantage Estimation in GRPO
Width Independent Bounds for the Local Lipschitz Constant of Deep Neural Networks at Random Initialization and after Lazy Training
WildActor: Unconstrained Identity-Preserving Video Generation
WildCat: Near-Linear Attention in Theory and Practice
WinDeskGround: A Benchmark for Robust GUI Grounding in Complex Multi-Window Desktop Environments
WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling
Winformer: Transcending Pairwise Similarity for Time-series Generation
WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points
WISE: World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation
With Argus Eyes: Assessing Retrieval Gaps via Uncertainty Scoring to Detect and Remedy Retrieval Blind Spots
WMVLM: Evaluating Diffusion Model Image Watermarking via Vision-Language Models
Words Towards Explainability: Caption Label-Free Learning via Dual Loop Agentic Time Series Captioning
Words & Weights: Streamlining Multi-Turn Interactions via Co-Adaptation
WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching
WorldComp2D: Spatio-semantic Representations of Object Identity and Location from Local Views
WorldCompass: Reinforcement Learning for Long-Horizon World Models
World Guidance: World Modeling in Condition Space for Action Generation
WorldMirror: Universal 3D World Reconstruction with Any-Prior Prompting
World-Model Inspired Emotion-aware Token Refinement for Training-Free Multimodal Emotion Recognition
World Models in Pieces: Structural Certification for General Agents
World Models in Pieces: Structural Certification of General Agents
WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling
World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

« prev Page 129/130 next »