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
179 results for “keyword:"model evaluation"” in ICLR 2027. 41 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Does Probe Conditioning Isolate Composition? An Information-Control Audit of Multi-Hop Evaluation▼1361.33
[PDF]When Better Routing Hides a Worse Prompt Update: Classifier-Conditioned Evaluation in Domain-Incremental Learning▲6.1k2.00
[PDF]OpticalBench: A Physics-Grounded Benchmark for Vision-Language Reasoning on Optical Experiments▼5.8k2.33
[PDF]Rating-Scale Drift in Automated Paper Reviewers after Domain-Specific Reinforcement Learning▼1.3k1.75
[PDF]Behavior and Self-Report Are Differentially Controllable: A Within-Model Causal Test in Ten Instruction-Tuned LLMs▼1.2k1.67
[PDF]Normal Fréchet Drift: Measuring Post-Training Departure from Pretrained Reference Manifold▲13k2.33
[PDF]Certifying What Your Generative Model Covers: Labeller-Conditional, Finite-Sample Mode-Recall Guarantees▲6.1k2.00
[PDF]What FID Hides: Detecting, Ranking, and Diagnosing Deviations in Generative Evaluation▼4.3k2.00
[PDF]RubricEdit-Bench: Benchmarking Complex Infographic Editing with Fine-Grained Diagnostic Evaluation▼4.3k2.00
[PDF]Search Finds Blind Spots: Stress-Testing Edit Reward Models Under Optimization Pressure▲3972.67
[PDF]MM-ImpBench: Measuring Multimodal Implicitness for Graded Evaluation of Vision–Language Models▼4.5k2.67
[PDF]Hypothesis Testing with Conditional Queries: Learnability and the Value of Interaction▼4.5k2.25