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
23 results for “keyword:"benchmarking and evaluation"” in ICLR 2027. 6 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]When LLMs Bid Rationally, and When They Don't: A Traceable Capability Split in Sealed-Bid Auctions▼1361.33
[PDF]FDE-Bench: Evaluating End-to-End Delivery from Underspecified Real-World Business Requests▼1.2k3.00
[PDF]Short-Moment Grounding in VideoLLMs: A Search-Space Limit Rather Than a Capability Limit▼4.5k2.67
[PDF]Position: Why AI For PHM May Never Have Its "ImageNet Moment" (and Why It Probably Shouldn't)▼1.2k1.67
[PDF]DuplexAct: Benchmarking Natural and Controllable Interaction Behaviors in Real-Time Full-Duplex Speech Systems▼4.3k2.00
[PDF]MOVI-Agent: Learning Queryable Software Models from Instructional Videos for Professional GUI Automation▼4.3k2.00
[PDF]TCSALGBENCH: BENCHMARKING AUTOMATED PROVING FOR RESEARCH-LEVEL THEORETICAL COMPUTER SCIENCE▲5.4k3.00
Withdrawn or desk rejected (6)
[PDF]CALIBRA-fMRI: Text for Signals Without Captions, Read by a Frozen Video-Language ModelWithdrawn2.00
[PDF]Bench2Drive-Plan: Bridging Privileged Planner and End-to-End Autonomous Driving under Shared ProtocolsWithdrawn1.50