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
11,760 results for “keyword:"models"” in ICLR 2027. 2,911 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Seeing More, Refusing Less: Two Mechanisms of Safety Erosion in Video Large Language Models▲3.0k2.33
[PDF]Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes▲3952.67
[PDF]BELT : BEhavior-Aligned Low-Rank KV Cache Compression with Learned Top-k Rank Allocation▲3.0k2.50
[PDF]What Should a World-Action Model Imagine? 4D Geometry as an Intermediate Representation for Control▲3.0k2.33
[PDF]SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation▼1.2k3.00
[PDF]DistillAlign: Coordinating Mode Covering and Mode Seeking in Autoregressive Video Distillation▲5103.00
[PDF]DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism▼4.3k2.00
[PDF]Auditing Optimization and Probabilistic Reliability in Frozen Vision-Language Representation Intervention▲3.9k1.67
[PDF]Beyond Holistic Representations: Dual Distribution Modeling for Test-Time Adaptation of Vision–Language Models▲3.0k2.33
[PDF]Render to Reason: Novel-View Semantic Prediction Improves Spatial Understanding in VLMs▼5.8k2.33
[PDF]Dataset Inference for Data Provenance and Privacy Auditing in Tabular Foundation Models▼4.3k2.00
[PDF]Situated Inference: Adapting Frozen Generative Speech Enhancement under Acoustic Distribution Shift▼5.8k2.33
[PDF]How Much to Trust the Eyes? Reliability-Guided Inference for Diffusion-Based Audio-Visual Speech Enhancement▲5103.00
[PDF]S2PD: Serial-to-Parallel Diffusion for Physically and Logically Consistent Video Generation▼5.8k2.33
Withdrawn or desk rejected (2,911)
[PDF]Mind Your Marginals: What Governs Fidelity in Linearized Hierarchical Optimal Transport?Desk rejected1.00
[PDF]A Window Is Not a Door: Mapping the Limits and Conditions of Knowledge Editing in Language ModelsDesk rejected2.33
[PDF]Propagation Is Not Decoration: When Physical Diffraction Carries Generative ComputationWithdrawn1.00
[PDF]When Is a Policy Violation a Model Failure? Policy Observability and Failure Attribution in Agent-Safety EvaluationDesk rejected