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
1,450 results for “keyword:"efficient"” in ICLR 2027. 350 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Compression-Induced Representational Dissonance: Anticipating and Suppressing Prediction Flips▼4.3k2.00
[PDF]ThunderWorld: Efficient World Model Generation with Selective FP4 Refinement via Native GEMMs▲3.0k2.33
[PDF]TreePC: Posterior-Consistent and Dependency Aware Decoding for Few-Step Diffusion Language Models▼4.3k2.00
[PDF]The Context of Theseus: Learning Selective Discrete Recurrence in Autoregressive Transformers▲6.1k2.00
[PDF]Beyond Sequential Scanning: Rethinking Visual State Propagation with Neighborhood State Interaction▼12k1.67
[PDF]Graphene: A Lightweight Transformer with Ising Mean-Field Refinement for Image Super-Resolution▼4.3k2.00
[PDF]Label-Free Steering: Compressing Test-Time Reinforcement Learning into Bias-Only Subspaces▲6.1k2.00
[PDF]Where Should Multimodal Test-Time Adaptation Capacity Live? A Controlled Study of Prompts and Residual Adapters▼1361.33
[PDF]Stable Geometry with Divergent Task Evidence for Efficient Long-Horizon Agent Compression▲3942.67
[PDF]Can we improve the performance of sparse attention in Grouped Query Experts without increasing experts?▲11.00
[PDF]DADP: Dynamic Activity-Dependent Pruning, A Reverse Hebbian-Inspired Structural Plasticity Method▼1361.33
[PDF]WAM-Diff2: Hierarchical AR-to-Diffusion Distillation for Highly Efficient Autonomous Driving VLA▼4.5k2.67
[PDF]VDGS: Improving Fidelity in Fast 3D Gaussian Splatting Training via Virtual Densification▲9.2k2.67
[PDF]A Light Bilevel Refinement Aligns Self-Supervised Representations for Stronger Task‑Specific Learning▲6.1k2.00
[PDF]SLaMES: Sensitivity-Guided Layerwise and Modality-Aware Expert Skipping for Efficient Inference in MoE MLLMs▲14k3.00
[PDF]Adaptation Is Not Retention: Reframing Continual Low-Rank Adaptation through Response Coordinates▼5.8k2.33
[PDF]Prune Once, Route per Turn: Constructing Persistent Visual Memory for Multi-Turn VLM Inference▼4.3k2.00
[PDF]Mixture of Outputs: Dynamic Decoding-Time Expertise Integration Between Large Language Models▼5.8k2.33
[PDF]Backpropagated Output Momentum: Relocating Optimizer History from Parameters to Task Space▲5103.00
Withdrawn or desk rejected (350)
[PDF]Mind Your Marginals: What Governs Fidelity in Linearized Hierarchical Optimal Transport?Desk rejected1.00
[PDF]CCQ: Compensation-Conditioned Quadratic Selection for Absorbing Skipped Experts in MoE LLMsWithdrawn2.00