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
11,753 results for “keyword:"models"” in ICLR 2027. 2,918 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]The Geometry of Refusal: Why Post-Hoc Safety is Fragile and Pretraining-Time Safety Persists▼1.2k1.67
[PDF]Fair Protocol Reveals Collapse of Medical Hallucination Detectors-and Two Detectors That Survive It▼1.2k1.67
[PDF]CapField-OPD: Learning Continuous Capability Fields via Joint-Anchored Multi-Teacher On-Policy Distillation for Flow Models
[PDF]Source or Prefix? Retaining What Helps the Target in Autoregressive 3D Medical Image Translation▼4.5k2.67
[PDF]After the Second Bell: Can Audio-Language Models Reason with Temporally Grounded Events?▼1.3k1.80
[PDF]TRACE: Novelty-Aware Spatiotemporal Evidence Retention for Streaming Video Understanding▼4.3k2.00
[PDF]Global Selection, Local Credit: Decoupled Topology Search for Adaptive Test-Time Scaling▲3.0k2.33
[PDF]Defence against the dark arts of distillation: Unsupervised discovery of subliminal learning▲3822.67
Withdrawn or desk rejected (2,918)
[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