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
56 results for “keyword:"token level"” in ICLR 2027. 5 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Family-Dependent Changes in Token-Level Prompt Dependence across Post-Training Checkpoints▲6.1k2.00
[PDF]Dr. OPD: Learning What to Follow for Optimal On-Policy Distillation of Large Language Models▲4012.67
[PDF]Token-Level Model Blending and AI-Generated Text Detection: Separating Model Switching from the Generation Loop▼1.2k1.67
[PDF]I Know That I Know Nothing: Narrowing the Knowledge-Output Gap in Large Language Models through Early High-Entropy Token Intervention▲3.0k2.33
[PDF]PaRa-OPD: Pathology Reward-Aware On-Policy Distillation via Token-Level Supervision Reweighting▼4.3k2.00
[PDF]Detection of Machine-Generated Text Using Fine-Tuned Transformers and a Compact Classifier Based on Token-Level Probabilistic Features▼4.3k2.00
[PDF]The Right Kind of Disagreement: Dynamic Collaborator Pruning for Speculative Token-Level Collaboration▼5.8k2.33
[PDF]Not All Tokens Are Equal: Entropy-Guided Credit Assignment for Direct Preference Optimization▼4.3k2.00
[PDF]Where Does RLVR Act? A Conserved Token Ledger for What Post-Training Adds to a Base Model▼4.3k2.00
[PDF]PeerCredit: Cross-Prompt Token Credit Assignment for Reinforcement Learning with Verifiable Rewards▲5.4k3.00
[PDF]MAX-Cache: Motion-Aligned Activation Reuse with Mismatch-Guided Selective Recomputation for Efficient Autoregressive Video Diffusion▲4012.67
[PDF]Where Should the Credit Lie? Attention-Induced Anchor-Bridge Attribution Policy Optimization for LLM Reasoning▼4.3k2.00
[PDF]How to Crack the Black Box of Emotion Generation? Interpretable Dual-Path Function Vectors for Token-Level Control▲6.1k2.00
[PDF]TraceStencil: Signed Temporal Neighborhoods for Token-Level Policy Optimization in RLVR▼1.2k1.67
Withdrawn or desk rejected (5)
[PDF]Endogenous Signals Fail to Allocate: Per-Token Compute Value Is Real, Transient, and Not LearnableWithdrawn1.33