ICLR 2026

5,370 papers

Top Research Directions

Research directions most represented by ICLR 2026 papers.

Scaling Language Model Reasoning Beyond Verifiable Rewards 88
Group Relative Policy Optimization for Large Reasoning Models 81
General Purpose Reasoning from Specialized Reinforcement Learning 79
Shifting from Static Benchmarks to Dynamic Evaluation 66
The Decoupling of Visual Perception and Logical Reasoning 51
Mitigating the Inference Latency of Chain of Thought 50
Activation Steering for Inference Time Model Control 41
Preventing Entropy Collapse in Reinforcement Learning 41
Inference Time Compute Scaling for Complex Reasoning 38
Reinforcement Learning with Verifiable Rewards 38
Autonomous Code Refinement via Execution Feedback Loops 38
Mitigating Computational Overthinking in Large Reasoning Models 37
Resolving Superposition with Sparse Autoencoders 34
Role Specialization in Multi-Agent Architectures 34
The Spatial Reasoning Bottleneck in Multimodal Models 33

Top Authors

AuthorPapers
Fei Huang20
Ge Zhang19
Jianye HAO19
Dahua Lin17
Junchi Yan17
Michael M. Bronstein17
Yejin Choi17
James Y. Zou16
James Zou16
Bo Han15
Jingren Zhou15
Min Zhang15
Tat-Seng Chua15
Caiming Xiong14
Dawn Song14
Pengfei Wan14
Yi Yang14
Jiaheng Liu13
Kai Chen13
Pengjun Xie13

Impactful Papers

Papers from ICLR 2026 that shaped research clusters.

Foundational

Papers

Title
Best-of-three-worlds Analysis for Dueling Bandits with Borda Winner
Better Bounds for the Distributed Experts Problem
Better Learning-Augmented Spanning Tree Algorithms via Metric Forest Completion
Better Together: Leveraging Unpaired Multimodal Data for Stronger Unimodal Models
Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning
Beyond Accuracy: Are Time Series Foundation Models Well-Calibrated?
Beyond Aggregation: Guiding Clients in Heterogeneous Federated Learning
Beyond a Million Tokens: Benchmarking and Enhancing Long-Term Memory in LLMs
BeyondBench: Contamination-Resistant Evaluation of Reasoning in Language Models
Beyond Binary Preferences: A Principled Framework for Reward Modeling with Ordinal Feedback
Beyond Binary Rewards: Training LMs to Reason About Their Uncertainty
Beyond Classification Accuracy: Neural-MedBench and the Need for Deeper Reasoning Benchmarks
Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning
Beyond Distributions: Geometric Action Control for Continuous Reinforcement Learning
Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs
Beyond Ensembles: Simulating All-Atom Protein Dynamics in a Learned Latent Space
Beyond Entity Correlations: Disentangling Event Causal Puzzles in Temporal Knowledge Graphs
Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Models
Beyond Grid-Locked Voxels: Neural Response Functions for Continuous Brain Encoding
Beyond Hearing: Learning Task-Agnostic ExG Representations from Earphones via Physiology-Informed Tokenization
Beyond In-Domain Detection: SpikeScore for Cross-Domain Hallucination Detection
Beyond Instance-Level Alignment: Dual-Level Optimal Transport for Audio-Text Retrieval
Beyond Length: Quantifying Long-Range Information for Long-Context LLM Pretraining Data
Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
Beyond Linear Processing: Dendritic Bilinear Integration in Spiking Neural Networks
Beyond Magic Words: Sharpness-Aware Prompt Evolving for Robust Large Language Models with TARE
Beyond Magnitude: Leveraging Direction of RLVR Updates for LLM Reasoning
Beyond Markovian Drifts: Action-Biased Geometric Walks with Memory for Personalized Summarization
Beyond Markovian: Reflective Exploration via Bayes-Adaptive RL for LLM Reasoning
Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes
Beyond Match Maximization and Fairness: Retention-Optimized Two-Sided Matching
Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy
Beyond Multi-Token Prediction: Pretraining LLMs with Future Summaries
Beyond Noisy-TVs: Noise-Robust Exploration Via Learning Progress Monitoring
Beyond Outliers: A Study of Optimizers Under Quantization
Beyond Pairwise: Empowering LLM Alignment With (Ranked) Choice Modeling
Beyond Pass@ 1: Self-Play with Variational Problem Synthesis Sustains RLVR
Beyond Penalization: Diffusion-based Out-of-Distribution Detection and Selective Regularization in Offline Reinforcement Learning
Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts
Beyond RAG vs. Long-Context: Learning Distraction-Aware Retrieval for Efficient Knowledge Grounding
Beyond Raw Detection Scores: Markov-Informed Calibration for Boosting Machine-Generated Text Detection
Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs
Beyond RLHF and NLHF: Population-Proportional Alignment under an Axiomatic Framework
Beyond Scattered Acceptance: Fast and Coherent Inference for DLMs via Longest Stable Prefixes
Beyond Sequential Reranking: Reranker-Guided Search Improves Reasoning Intensive Retrieval
Beyond Short Steps in Frank-Wolfe Algorithms
Beyond Simple Graphs: Neural Multi-Objective Routing on Multigraphs
Beyond Skeletons: Learning Animation Directly from Driving Videos with Same2X Training Strategy
Beyond Softmax and Entropy: Convergence Rates of Policy Gradients with $\boldsymbol{f}$-SoftArgmax Parameterization $\&$ Coupled Regularization
Beyond Spectra: Eigenvector Overlaps in Loss Geometry

« prev Page 10/108 next »