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
11,758 results for “keyword:"models"” in ICLR 2027. 2,913 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Reasoning or Retrieval? Diagnosing LLM Failures on Temporal Knowledge Graphs with Ground-Truth Evidence▼5.9k2.33
[PDF]Think Twice Before Calling Tools: How Pre-Call Reasoning in SFT Shapes Tool-Use Retention under RL▼5.9k2.33
[PDF]Hesitation Has a Geometry: Entropy-Trained Hyperbolic Probes for Sparse Activation Steering▼4.3k2.00
[PDF]Query Complexity of Forward-Only Hebbian Learning: Theory and a Native Fit for Diffusion Models▼4.3k2.00
[PDF]Budgeted Multimodal Evidence Selection for Long-Video Understanding via Knapsack Optimization▲3.0k2.33
[PDF]From Pixels, Without Pre-training: Joint Generative and Self-Supervised Representation Learning in One Model▲9.2k2.67
[PDF]Plottability Is Not Identifiability: Auditing the Retraining Stability of KAN Explanations▼1361.33
[PDF]Sample the Order, Not the Token: Test-Time Scaling for Masked Diffusion Language Models▲3912.67
[PDF]FMD-IMHF: Foundation Model Distillation and Hypergraph Fusion for Neurodegenerative Disease Diagnosis with Incomplete Multimodal Data▲5.4k3.00
[PDF]When Are Foundation Models Good Distillation Teachers? A Controlled Study of Teacher Selection in ECG▼4.3k2.00
Withdrawn or desk rejected (2,913)
[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