Optimization & Theory

LightRR: A Lightweight Network for Single Image Reflection Removal CVPR 2026 Linear-Time and Constant-Memory Text Embeddings Based on Recurrent Language Models ACL 2026 LinVideo: A Post-Training Framework towards O(n) Attention in Efficient Video Generation CVPR 2026 LiteVGGT: Boosting Vanilla VGGT via Geometry-aware Cached Token Merging CVPR 2026 Lizard: An Efficient Linearization Framework for Large Language Models ACL 2026 Logit-Margin Repulsion for Backdoor Defense CVPR 2026 LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification ACL 2026 Look Within or Beyond? A Theoretical Comparison Between Parameter-Efficient and Full Fine-Tuning ACL 2026 LRDUN: A Low-Rank Deep Unfolding Network for Efficient Spectral Compressive Imaging CVPR 2026 LS-ViT: Least-Squares Hessian Based Block Reconstruction for Low-Bit Post-Training Quantization of Vision Transformers CVPR 2026 Machine Unlearning via Adaptive Gradient Reweighting and Multi-stage Objective Optimization CVPR 2026 Make it SING: Analyzing Semantic Invariants in Classifiers CVPR 2026 Making Large Language Models Efficient Dense Retrievers ACL 2026 ManifoldNeuS: Manifold-aware View Optimizability for Pose-Free Neural Surface Reconstruction CVPR 2026 Mapping Networks CVPR 2026 MATCH: Modulating Attention via In-Context Retrieval for Long-Context Transformers ACL 2026 Mechanistic Interpretability Should Prioritize Feature Consistency in Sparse Autoencoders ACL 2026 Memory-Efficient Transfer Learning with Fading Side Networks via Masked Dual Path Distillation CVPR 2026 Merge3D: Efficient 3D Multimodal LLMs via Joint 2D-3D Token Merging CVPR 2026 Mitigating The Distribution Shift of Diffusion-based Dataset Distillation CVPR 2026 MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context Reasoning ACL 2026 Mixture of Heterogeneous Grouped Experts for Language Modeling ACL 2026 MM-SeR: Multimodal Self-Refinement for Lightweight Image Captioning CVPR 2026 Model Merging is Secretly Certifiable: Non-Vacuous Generalisation Bounds for Low-Shot Learning UAI 2026 Mostly Text, Smart Visuals: Asymmetric Text-Visual Pruning for Large Vision-Language Models CVPR 2026