NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
628 results for “keyword:"understanding"” in ICLR 2027. 278 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]IR³: Index Once, Retrieve, Restore, and Reduce for Multi-Query Long-Video Understanding▲3.0k2.33
[PDF]Learning to Reason Efficiently in Videos: Reinforcement Learning with Short- and Long-Term Rewards▼1.2k1.67
[PDF]MARD: Multi-level Adaptive Representation Distillation across Heterogeneous Architectures for Egocentric Action Recognition▼4.3k2.00
[PDF]MOBA-VL: Event-Localized Multi-Turn Reinforcement Learning for Real-Time MOBA Commentary▲432.75
[PDF]Can We Read the Mind of an Audio LLM? A Verbalizable, Multilingual Middle-Layer Workspace▲432.75
[PDF]Omni-Diffusion-Distill: Few-Step Distillation of Unified Multimodal Diffusion Large Language Models▼1.2k3.00
[PDF]Counterfactual Answer Prediction Training: Improving LLMs' Causal Understanding via Input-Perturbation Difference Prediction▼1.2k1.67
[PDF]Seeing Risk Before It Happens: Risk-Augmented World Model for Anticipatory Driving Risk Understanding▼5.8k2.33
[PDF]When Should the Decision Change? Counterfactual Probing of Vision-Language Models for Autonomous Driving▼4.3k2.00
[PDF]STRAND: Benchmarking and Improving Object-Centric Spatio-Temporal Monitoring in Video Large Language Models▲5103.00
[PDF]SEGA: Multi-View 3D Instance Segmentation with Geometry-Guided Association for Indoor Environments▲3952.67
[PDF]From Surface to Depth: Towards Cognitive Appraisal Reasoning in Multimodal Emotion Understanding▲4.6k3.00
[PDF]Prediction Can Benefit Understanding: Visual Prediction as Complementary Supervision for Video LLMs▲9.2k2.67
[PDF]CrossTimeBench: Benchmarking Cross-Temporal Video Reasoning in Complex Multi-Event Scenarios▼1.3k1.75
[PDF]LEMMO: A Large Electromagnetic Model Leveraging Pretrained LLMs for Million-Sample Signal Understanding▼4.3k2.00
[PDF]Encoded but Unread: Locating and Recovering Within-Event Motion in Video Language Models▼12k1.67
[PDF]VEOcc++: Learning Persistent Voxel Feature Maps for Embodied Online Semantic Occupancy Mapping and Offline Refinement▼1.2k3.00
[PDF]Partition and Select: Efficient Training-Free Keyframe Selection for Long-Video Understanding▼4.3k2.00
[PDF]Exploiting Video Temporal Capability as an Attack Pathway in Multimodal Large Language Models▼4.3k2.00
[PDF]MuseFACET: Benchmarking Trustworthy and Controllable Image-to-Text Generation for Musuem Collection Interpretation▲5.4k3.00
[PDF]Can LLMs Think Like Consumers? Benchmarking Crowd-Level Reaction Reconstruction with ConsumerSimBench▲6.1k2.00
[PDF]Cross-Feature Guided Reinforcement Learning for Unified Multimodal Understanding and Generation▲3.0k2.33
Withdrawn or desk rejected (278)
[PDF]Beyond Local Cycles: Long-Horizon Repetitive Action Counting via Multi-Scale Temporal Representation LearningDesk rejected
[PDF]Beyond Layers: Position-Resolved Gradient Conflict and Position-Aware Modulation for Unified Multimodal ModelsWithdrawn2.00
[PDF]TFAG-OVD: Training-Free Attribute-Guided Open-Vocabulary Detection via LLM Semantic CompositionWithdrawn1.67
[PDF]Now You See It, Now You Don’t: Rethinking visual token allocation across Vision Language Model layersWithdrawn1.33
[PDF]LongVU-TTT: Causal Test-Time Training for Visual Resampling in Long Video UnderstandingWithdrawn1.75
[PDF]VideoEvoHarness: Self Evolving Agent Harness from Unlabeled Videos for Video UnderstandingWithdrawn1.75