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
Predicting evolutionary rate as a pretraining task improves genome language model representations
Predicting Future Utility: Global Combinatorial Optimization for Task-Agnostic KV Cache Eviction
Predicting Large Model Test Losses with a Noisy Quadratic System
Predicting the Emergence of Induction Heads in Language Model Pretraining
Predicting the Order of Upcoming Tokens Improves Language Modeling
Predicting What Matters: Robust Generalist Robot Policy Learning via Future Semantic Mask
Prediction-Powered Adaptive Inference with Pretrained AI Models for Contextual Bandits
Prediction-Powered Risk Monitoring of Deployed Models for Detecting Harmful Distribution Shifts
Predictive Prefetching for Retrieval-Augmented Generation
Predictive variational inference: Learn the predictively optimal posterior distribution
Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision
Preference-Calibrated Optimization with Score-Level Distribution Alignment for Text-to-Image Diffusion Model Unlearning
Preference-Enhanced Reinforcement Learning for Pluralistic Image Inpainting
Preference Goal Tuning: Post-Training as Latent Control for Frozen Policies
Preference-Modulated Structural Attention for Multi-Objective Combinatorial Optimization
Prefix-Cache-Aware Data Reordering for LLM-Augmented Database Analytics
Prescriptive Scaling Reveals the Evolution of Language Model Capabilities
Preserve-Then-Quantize: Balancing Rank Budgets for Quantization Error Reconstruction in LLMs
Preserving Plasticity in Continual Learning via Dynamical Isometry
Pressure Reveals Character: Behavioural Alignment Evaluation at Depth
Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning
PretrainZero: Reinforcement Active Learning on Pretraining Data
Primal-Spectral Generative Modeling: Fast Analytical Generation via Pseudoinverse Lévy Inversion
PRIM: Cooperative Dynamic Token Compression for Efficient Large Multimodal Models
PRIM:Cooperative Dynamic Token Compression for Efficient Large Multimodal Models
Principled Confidence Estimation for Deep Computed Tomography
Principled RL for Flow Matching Emerges from the Chunk-level Policy Optimization
Principled SVD-based Delta Compression via Quantization Error Minimization
Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation
Principle-Evolvable Scientific Discovery via Uncertainty Minimization
Prior Diffusiveness and Regret in the Linear-Gaussian Bandit
Prioritized Model Experience Replay
Prioritize the Process, Not Just the Outcome: Rewarding Latent Thought Trajectories Improves Reasoning in Looped Language Models
Priority-Aware Shapley Value
PRISM: Demystifying Retention and Interaction in Mid-Training
PRISM: Distribution-free Adaptive Computation of Matrix Functions for Accelerating Neural Network Training
PRISM: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models
PRISM: Gauge-Invariant Tangent-Space Differentially Private LoRA
PRISM: Learning Realistic Depth via Physics-Grounded Noise Disentanglement with Semantic-Geometric Collaboration
Prism-MoE: Efficient Dense-to-MoE Conversion for Visual Autoregressive Generation
PRISM: Parallel Residual Iterative Sequence Model
PRISM: Perception Reasoning Interleaved for Sequential Decision Making.
Prism: Spectral-Aware Block-Sparse Attention
PRISM: Synergizing Vision Foundation Models via Self-organized Expert Specialization
PRISM: Training-Free Video Anomaly Detection via Intrinsic Statistical Modeling
PrivAct: Internalizing Contextual Privacy Preservation via Multi-Agent Preference Training
Privacy Amplification in Differentially Private Zeroth-Order Optimization with Hidden States
Privacy-Aware Data Integration for Enhanced Quantile Inference under Heterogeneity
Privacy-Aware Video Anomaly Detection through Orthogonal Subspace Projection
Privasis: Synthesizing the Largest "Public" Private Dataset from Scratch

« prev Page 87/130 next »