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
11,752 results for “keyword:"models"” in ICLR 2027. 2,919 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]From Independent to Context-Aware Inference via Tabular Foundation Models for Energy Prediction on 3D Atomic Structures▼1.2k1.67
[PDF]DuplexOmni: Real-Time Listening, Seeing, Thinking, and Speaking for Full-Duplex Interaction▲5.4k3.00
[PDF]Harnessing Domain Specialists in Multimodal Mixture-of-Experts for Efficient Adaptation▲5093.00
[PDF]Watt Does a Benchmark Point Cost? Measuring the Energy Efficiency of LLM Capability under Production Serving▲3.0k2.33
[PDF]CG-DIFF: CAUSAL-GRAPH-GUIDED DIFFUSION MODELS FOR TABULAR IMPUTATION UNDER SELF-MASKING MNAR▼1.2k1.67
[PDF]Harnessing Multimodal Large Language Models for Training-free Human-object Interaction Detection▲3.0k2.33
[PDF]PLAN-Bench: Can Existing VLMs Go Well Beyond Perception to Reason and Plan in Embodied Worlds?▼1.2k1.67
Withdrawn or desk rejected (2,919)
[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