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11,760 results for “keyword:"models"” in ICLR 2027. 2,911 withdrawn or desk rejected listed apart.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]When Is an Attention Head a Computational Unit? Shared Circuits Despite Orthogonal Representations▲5103.00
[PDF]Fine-Tuned Judgment Lives in the Weights: In-Context Examples Help Untrained Models and Override Trained Ones▼1361.33
[PDF]Correctable Means Corruptible: The Reasoning Chain of Vision-Language-Action Policies Under Adaptive Attack▼4.3k2.00
[PDF]VLA-Hallucination: Diagnosing and Mitigating Hallucination Failures in Vision-Language-Action Models▲5.4k3.00
[PDF]SIRRA: Matched Evaluation of Standing-Instruction Reapplication after Temporary Non-Applicability▼4.3k2.00
[PDF]Branch-JEPA: Learning Dynamics-Aware State Representations through Dual-Branch Patch Routing▼1361.33
[PDF]OmniCens: Self-Improving Distillation for Multimodal Classification of Millions of Profiles▲11.00
[PDF]BiGenCap: Joint Bitemporal Generative Modeling via Reciprocal Prediction for Remote Sensing Image Change Captioning▼5.8k2.33
[PDF]LatentSpec: Two-Stage Thinker–Drafter Parallel Drafting with Shared Latent Reasoning for Speculative Decoding▲9.2k2.67
[PDF]ProtoLoRA: Visual-Prototype-Guided LoRA Backdoor Attack against Vision–Language Models▼5.8k2.33
[PDF]A Harmful Request Is Still a Request: Benign Fine-Tuning Rewrites How Aligned Models Begin to Answer▲3952.67
[PDF]Towards All-Atom Protein World Models: Learning and Diagnosing Predictive Latent Representations▼4.3k2.00
[PDF]TrustRoboReward: Preference-Ordered Isotonic Score Editing for Multi-Paradigm Robot Reward Models▼4.3k2.00
[PDF]Diffusion Additive Models: Interpretable Generative Prediction through Feature-Wise Contributions▼1361.33
[PDF]Not Proven Is Not Proven-Not: Language Models Read the “Not” in a Question as Negation as Failure▼4.3k2.00
[PDF]A Jailbreak Success Coordinate Outside the Refusal Subspace: Decodable Before Generation, Reachable from the Input▼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Withdrawn or desk rejected (2,911)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