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
40 results for “keyword:"draft"” in ICLR 2027. 3 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Beyond the Marginals: Dependence and Parallel Decoding in Masked Diffusion Language Models▼1.2k1.67
[PDF]Frequencies Rank, Probabilities Allocate: Calibrated Draft Trees for Drafter-Free Speculative Decoding▲4032.67
[PDF]When to Trust the Past: Reliability-Calibrated Draft-Tree Orchestration for Speculative Decoding▲4032.67
[PDF]ReDraft: Your Verifier Is Secretly a Depth Supervisor for Recurrent Speculative Decoding▲5103.00
[PDF]Fork and Regenerate: MTP-Guided Early Forking and Suffix Regeneration for Block Diffusion Speculative Decoding▲3.0k2.33
[PDF]Co-Evolving Trees and Control Policies: Bridging Posterior Optimality and Online Efficiency in Speculative Decoding▼5.8k2.33
[PDF]Calibrate Before You Branch: Feedback-Calibrated, Cost-Aware Trees for Block-Diffusion Speculative Decoding▼4.3k2.00
[PDF]On the SPOT: Enhancing Speculative Decoding for Diffusion Language Models via Commit-Order Prediction▼2433.33
[PDF]DSpark-OPD: Acceptance-Aligned On-Policy Distillation for Semi-Autoregressive Speculative Drafting▼1.2k1.67
[PDF]When to Evict, Not What to Keep: Draft-Guided Eviction for Training-Free KV-Cache Compression▼1.2k3.00
Withdrawn or desk rejected (3)
[PDF]Separating marginal error from recoverable dependence in parallel-head speculative decoding for large language modelsDesk rejected2.33