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
86 results for “keyword:"adaptive computation"” in ICLR 2027. 21 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Effective Does Not Mean Useful: Conditional Functional Substitutability for Redundancy and Scaling in Transformers▼5.8k2.33
[PDF]Optimal Depth of Neural Networks: Resource-Correct Stopping with Bounded Logit Increments▼1.2k1.67
[PDF]Listen First, Look as Needed: Uncertainty-Aware Asymmetric Routing for Audio-Visual Recognition▼1351.33
[PDF]ARCS: APPROXIMATION-RISK-AWARE COMPUTATION SCHEDULING FOR FEATURE FORECASTING IN DIFFUSION TRANSFORMERS▲4012.67
[PDF]Residual Geometry Separates State Motion from Predictive Change in Weight-Tied Iterative Refinement▼1.2k1.67
[PDF]ReDraft: Your Verifier Is Secretly a Depth Supervisor for Recurrent Speculative Decoding▲5103.00
[PDF]ROUTING SHUFFLES DO NOT IDENTIFY ADAPTIVE BENEFIT: THEORY AND A CONTROLLED ACM-GCN AUDIT▼1351.33
[PDF]SAFER: WHEN IS ACQUIRING FORENSIC EVIDENCE WORTH THE COST IN AGENTIC AI-GENERATED VIDEO DETECTION?▲6.1k2.00
[PDF]StopTheDream: Outcome-Grounded Candidate Comparison for Adaptive Search in World-Model Planning▲4012.67
[PDF]DGTE: Disagreement-Gated Test-Time Evolution of Roles and Topologies in Multi-Agent Systems▼1351.33
[PDF]Where to Speculate: Cross-Request Survival-Value Allocation for Batched Speculative Decoding▼1.2k1.67
[PDF]Learning Whether to Learn: Prospective Counterfactual Test-Time Learning for Language Models▼4.3k2.00
[PDF]High-Load Budgeted Categorization of Customer Care Calls: An Encoder-LLM Cascade Solution▼1351.33
[PDF]HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment▼1351.33
[PDF]Iterate, Don't Stack: Grounded Recursive Transformers for Generalizable Novel View Synthesis▲5.4k3.00
Withdrawn or desk rejected (21)
[PDF]Same Answer, Different Futures: Judging Latent States by Their Continuations in Looped TransformersDesk rejected
[PDF]Endogenous Signals Fail to Allocate: Per-Token Compute Value Is Real, Transient, and Not LearnableWithdrawn1.33
[PDF]RAJNI: Importance-Aware, Dispersion-Scheduled Training-Free Token Pruning for Vision TransformersWithdrawn1.67
[PDF]TreeSpark: From Acceptance Scores to Useful Work in Conditional Speculative DecodingWithdrawn1.33