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
180 results for “keyword:"llm inference"” in ICLR 2027. 24 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]ONE VOCABULARY PASS, MANY REQUESTS: A STREAM-ONCE OUTPUT-HEAD KERNEL FOR CONCURRENT LLM DECODING▼4.3k2.00
[PDF]Frequencies Rank, Probabilities Allocate: Calibrated Draft Trees for Drafter-Free Speculative Decoding▲3952.67
[PDF]When to Wait Before Fan-Out: Warm-Up, Verification, and Residency in Cache-Aware LLM Evaluation▲9.2k2.67
[PDF]DStreamer: Acceptance-Aligned Diffusion Streaming for High-Throughput Serving and Online RLVR▲3952.67
[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]A RIGOR-MATCHED AUDIT OF PERIODIC-STEP LAYER SKIPPING FOR EFFICIENT LLM INFERENCE: CONFLAYERS VERSUS SWIFT, WITH A SUPPLEMENTAL ANALYSIS OF TRAINED ROUTING ALTERNATIVES▲11.00
[PDF]Co-Evolving Trees and Control Policies: Bridging Posterior Optimality and Online Efficiency in Speculative Decoding▼5.8k2.33
[PDF]WASP: Weighted Asymmetric Sparse Attention to Accelerate both Prefilling and Decoding in Long-context Models▲3.0k2.33
[PDF]Select Twice, Freeze Once: Jacobian-Based KV Cache Selection for Multi-Turn and Long-Context LLM Serving▼1.2k1.67
[PDF]Keys Decide Attention, Values Decide Contribution: Value-Gated KV Cache Eviction at 20× Compression▼1361.33
Withdrawn or desk rejected (24)
[PDF]AccelMark-Agent: Benchmarking Inference-Optimization Agents Across Accelerators and WorkloadsWithdrawn1.67
[PDF]ForeCache: Efficient LLM-Based Multi-Agent Systems with Intent-Conditioned Pruning and Amortized KV ReuseWithdrawn2.00
[PDF]CSD: Confidence- and Stability-Guided Speculative Decoding in Large Language ModelsWithdrawn1.33
[PDF]SHIM: Scaling Homomorphic Generative Transformer Inference via Multi-Party CollaborationWithdrawn1.00