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
Towards A Generative Protein Evolution Machine with DPLM-Evo
Towards a Holistic Understanding of Selection Bias for Causal Effect Identification
Towards a Science of AI Agent Reliability
Towards Atoms of Large Language Models
Towards a Unified Generative Model for Scarce Time Series with Domain Experts
Towards Budget-Aware Agents: Do LLM Agents Know What They Will Spend?
Toward Scalable and Valid Conditional Independence Testing with Spectral Representations
Towards Cold-Start Drafting and Continual Refining: A Value-Driven Memory Approach with Application to NPU Kernel Synthesis
Towards Complete Multi-Agent Coordination Policy Learning via Denoising Maximum Entropy Optimization
Towards Completeness in Causal Discovery from Soft Interventions with Known Targets
Towards Context-Invariant Safety Alignment for Large Language Models
Towards Diffeomorphism-Equivariant Neural Networks via Canonicalization
Towards Disentangled Preference Optimization Dynamics: Suppress the Loser, Preserve the Winner
Towards Diverse Scientific Hypothesis Search with Large Language Models
Towards Docking-oriented De Novo Ligand Design via Gradient Inversion
Towards Effective Waste Segmentation for Automated Waste Recycling in Cluttered Background
Towards Efficient and Expressive Offline RL via Flow-Anchored Noise-conditioned Q-Learning
Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression
Towards Efficient LLMs Annealing with Principled Sample Selection
Towards Execution-Grounded Automated AI Research
Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation
Towards Fine-Grained Robustness: Attention-Guided Test-Time Prompt Tuning for Vision-Language Models
Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy
Towards Fully Parameter-Free Stochastic Optimization: Grid Search with Self-Bounding Analysis
Towards Functional Correctness of Code Models with Selective Generation
Towards Functional Correctness of Large Code Models with Selective Generation
Towards Generative Graph Matching for Graph Edit Distance Computation
Towards Hierarchy–Uniformity Equilibrium: Recovering Semantic Depth in Hypergraph Contrastive Learning
Towards High-Fidelity CAD Generation via LLM-Driven Program Generation and Text-Based B-Rep Primitive Grounding
Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs
Towards Multimodal Large Language Models with Both Training and Inference Efficiency
Towards One-for-All Anomaly Detection for Tabular Data
Towards One-to-Many Temporal Grounding
Towards On-Policy SFT: Distribution Discriminant Theory and its Applications in LLM Training
Towards Optimal Robustness in Learning-Augmented Paging
Towards Parameter-Free Temporal Difference Learning
Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization
Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Models
Towards Professional-Grade Financial Agents: Benchmarking, Tooling, and Structured Reasoning
Towards Realistic Lifelong Re-identification: Identity Recurrence with Changing Clothes
Towards Reliable Marking and Verification of AI-Generated Text via Geometry-aware Sentence-level Watermarking
Towards Resource-Efficient LLMs: End-to-End Energy Accounting of Distillation Pipelines
Towards Rule-Based Knowledge Sharing in Federated Learning
Towards Scalable and Consistent 3D Editing
Towards Seed-Robust Safety Alignment in Text-to-Image Models
Towards Solving the Gilbert-Pollak Conjecture via Large Language Models
Towards Spectroscopy: Susceptibility Clusters in Language Models
Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions
Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer
Towards Sub-Second Molecular Docking as a Structural Primitive: A Quantized Consistency Diffusion Framework

« prev Page 118/130 next »