inference acceleration

Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Acceleration ICCV 2025 Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Models AAAI 2025 FOLDER: Accelerating Multi-Modal Large Language Models with Enhanced Performance ICCV 2025 FractalLLM: Lossless Self-Speculative Decoding with Layer Embedded Self-Compression EMNLP 2025 FREE: Fast and Robust Vision Language Models with Early Exits ACL 2025 FR-Spec: Accelerating Large-Vocabulary Language Models via Frequency-Ranked Speculative Sampling ACL 2025 “Give Me BF16 or Give Me Death”? Accuracy-Performance Trade-Offs in LLM Quantization ACL 2025 GQSA: Group Quantization and Sparsity for Accelerating Large Language Model Inference IJCNLP 2025 Grouped Speculative Decoding for Autoregressive Image Generation ICCV 2025 Growing a Twig to Accelerate Large Vision-Language Models ICCV 2025 HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference ACL 2025 Hierarchical Speculative Decoding with Dynamic Window NAACL 2025 HookMoE: A learnable performance compensation strategy of Mixture-of-Experts for LLM inference acceleration EMNLP 2025 IAM: Efficient Inference through Attention Mapping between Different-scale LLMs ACL 2025 ImagePiece: Content-aware Re-tokenization for Efficient Image Recognition AAAI 2025 Inference-Time Diffusion Model Distillation ICCV 2025 Inverse Bridge Matching Distillation ICML 2025 Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment ICLR 2025 KVCrush: Key Value Cache size-reduction using similarity in head-behaviour ACML 2025 LAVa: Layer-wise KV Cache Eviction with Dynamic Budget Allocation EMNLP 2025 LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers AAAI 2025 LazyMAR: Accelerating Masked Autoregressive Models via Feature Caching ICCV 2025 LeanK: Learnable K Cache Channel Pruning for Efficient Decoding EMNLP 2025 Lifting the Veil on Visual Information Flow in MLLMs: Unlocking Pathways to Faster Inference CVPR 2025 Lossless Acceleration of Large Language Models with Hierarchical Drafting based on Temporal Locality in Speculative Decoding NAACL 2025