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
452 results for “keyword:"efficiency"” in ICLR 2027. 94 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Watt Does a Benchmark Point Cost? Measuring the Energy Efficiency of LLM Capability under Production Serving▲3.0k2.33
[PDF]Termination Calibration Recovers Answers That Reasoning Models Compute but Never Express▼4.5k2.67
[PDF]MuLA: One Prefill, Many Agents via Efficient KV Cache Reuse in Multi-LoRA Agents Systems▲5103.00
[PDF]RATIO: Reasoning Analysis and Token-level Inference Optimization for Quantized Reasoning Models▼4.3k2.00
[PDF]PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding▼1.2k3.00
[PDF]TEMPORALLY CONSISTENT ROUTING: A TRAINING- TIME OBJECTIVE FOR CACHE-EFFICIENT MIXTURE-OF- EXPERTS▲6.1k2.00
[PDF]Speculative Pipeline Decoding: Higher-Accuracy Drafting with Hidden Latency via Pipeline Parallelism▲3952.67
[PDF]SIGMA: Label-Efficient Training of Language Models to Follow Changing Structured Guidance▼1.2k1.67
[PDF]RelFed: Relative-Representation Federation of Heterogeneous Small Language Models for Multi-Site AIoT Sensing▼4.3k2.00
[PDF]SwiftAD: Visibility-Guided Hierarchical Token Pruning for Efficient Zero-Shot 3D Anomaly Detection▼5.8k2.33
Withdrawn or desk rejected (94)
[PDF]Matched Counterfactual Attribution: A Controlled Evaluation of Data Selection for Video Instruction TuningDesk rejected2.00
[PDF]Less Data, Faster Convergence: Goal-Driven Data Optimization for Multimodal Instruction TuningDesk rejected1.67
[PDF]The Normalization Landscape in Continual Learning: Resolving Moving-Average Breakdown and Subspace Interference for Buffer-Free Edge IntelligenceWithdrawn
[PDF]Cross-Model KV Cache Construction in LLM Familiies: Residual-Stream Mapping Without Matched Head GeometryDesk rejected