Model Compression

VLA-Mark: A cross modal watermark for large vision-language alignment models EMNLP 2025 VLTP: Vision-Language Guided Token Pruning for Task-Oriented Segmentation WACV 2025 VoCo-LLaMA: Towards Vision Compression with Large Language Models CVPR 2025 Wanda++: Pruning Large Language Models via Regional Gradients ACL 2025 Watermarking One for All: A Robust Watermarking Scheme Against Partial Image Theft CVPR 2025 Watermarking with Low-Entropy POS-Guided Token Partitioning and Z-Score-Driven Dynamic Bias for Large Language Models EMNLP 2025 Weight-Aware Activation Sparsity with Constrained Bayesian Optimization Scheduling for Large Language Models EMNLP 2025 WET: Overcoming Paraphrasing Vulnerabilities in Embeddings-as-a-Service with Linear Transformation Watermarks ACL 2025 When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoning EMNLP 2025 Which Retain Set Matters for LLM Unlearning? A Case Study on Entity Unlearning ACL 2025 Who Taught You That? Tracing Teachers in Model Distillation ACL 2025 Why Do Some Inputs Break Low-Bit LLM Quantization? EMNLP 2025 WINS: Winograd Structured Pruning for Fast Winograd Convolution ICCV 2025 Word Salad Chopper: Reasoning Models Waste A Ton Of Decoding Budget On Useless Repetitions, Self-Knowingly EMNLP 2025 XQuant: Achieving Ultra-Low Bit KV Cache Quantization with Cross-Layer Compression EMNLP 2025 You Only Prune Once: Designing Calibration-Free Model Compression With Policy Learning ICLR 2025 Zero-Shot Cross-Lingual Transfer using Prefix-Based Adaptation EMNLP 2025 Zeroth-Order Fine-Tuning of LLMs with Transferable Static Sparsity ICLR 2025 zFLoRA: Zero-Latency Fused Low-Rank Adapters EMNLP 2025 ZigZagKV: Dynamic KV Cache Compression for Long-context Modeling based on Layer Uncertainty COLING 2025 ZIP: An Efficient Zeroth-order Prompt Tuning for Black-box Vision-Language Models ICLR 2025 $C^2M^3$: Cycle-Consistent Multi-Model Merging NeurIPS 2024 Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design NeurIPS 2024 3-in-1: 2D Rotary Adaptation for Efficient Finetuning, Efficient Batching and Composability NeurIPS 2024 Accurate Retraining-free Pruning for Pretrained Encoder-based Language Models ICLR 2024