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
2,349 results for “keyword:"evaluation"” in ICLR 2027. 571 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]CompulsionBench: A Benchmark for Auditing Compulsion-Risk Proxies in Sequential Recommendation▼1.2k1.67
[PDF]Capability Detection: Monitoring Hidden and Emergent Capabilities in Tool-Using Agents▲3.0k2.33
[PDF]What Does a Training Token Buy? Contextual Information Gain Beyond Perplexity Across Open Model Suites▼1.2k1.67
[PDF]Measure First: The Noise Floor of an Autonomous Research Loop Dominates Its Accept Rule▼5.8k2.33
[PDF]On the Fragility of Self-Improving Agents: Variance, Task Order, and Underspecification▼4.5k2.67
[PDF]Beyond Aggregate Scores: Revealing Conditional Model Behavior in Chest X-Ray Nodule Detection▲3.0k2.33
[PDF]When a Reliability Coefficient Orders References Backwards: Identifiability, a Calibrated Test, and Design Rules for the Human References of LLM Judges▼5.8k2.33
[PDF]Acting Through Others: Evaluating Agent Competence and Safety in Socially Distributed Environments▼1.2k3.00
[PDF]Causal Consequence Invariance: Do LLM Causal Predictions Depend on Downstream Preferences?▼4.3k2.00
[PDF]Correlation Does Not Certify Reliable Evaluation: Identifiability Limits of World-Model-Based Robot Policy Selection▼4.3k2.00
[PDF]How Much Do Antibody Benchmarks Depend on Related Training Examples? A Cross-Benchmark Audit▼11k1.80
[PDF]Right Destination, Wrong Journey: Synthetic Counseling Dialogues Miss Human Interactional Dynamics▼5.8k2.33
[PDF]The Floor That Barely Falls: Untrained Encoders as the Operative Null for Concept-Discovery Metrics▲11.00
[PDF]Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents▼5.8k2.33
[PDF]When Aggregate AUROC Ranks the Shortcut: Trivial-Axis Headroom in Vision-Language Hallucination Detection▲5103.00
[PDF]The Decomposition Tax: LLM Pipelines Lose Up to 40 Accuracy Points at Their Own Interfaces▼4.3k2.00
[PDF]When Evidence is Missing: Benchmarking Hallucination and Abstention from Incomplete Observations▼1.2k1.67
[PDF]A quoted answer overrides the rule that designates it: what an instruction-compliance score measures depends on how the rule names its answer▲11.00
Withdrawn or desk rejected (571)
[PDF]Machine Love: Emergent Bonding Behavior in Large Language Models and Its Implications for AI SafetyDesk rejected
[PDF]What a 3D Reconstruction Benchmark Can and Cannot Tell About Geometric Foundation ModelsWithdrawn2.00
[PDF]From Association to Decision Value: Testing Paired Instruction Dependence for VLA Corpus SelectionWithdrawn1.40