Model Compression

MorphMark: Flexible Adaptive Watermarking for Large Language Models ACL 2025 MosaicDiff: Training-free Structural Pruning for Diffusion Model Acceleration Reflecting Pretraining Dynamics ICCV 2025 MoST: Efficient Monarch Sparse Tuning for 3D Representation Learning CVPR 2025 MPQ-DM: Mixed Precision Quantization for Extremely Low Bit Diffusion Models AAAI 2025 MSQ: Memory-Efficient Bit Sparsification Quantization ICCV 2025 MTL-LoRA: Low-Rank Adaptation for Multi-Task Learning AAAI 2025 Multi-Modality Expansion and Retention for LLMs through Parameter Merging and Decoupling ACL 2025 Multimodal Promptable Token Merging for Diffusion Models AAAI 2025 Multi-Task Pre-Finetuning of Lightweight Transformer Encoders for Text Classification and NER EMNLP 2025 Neural Collapse Inspired Knowledge Distillation AAAI 2025 Neural Parameter Search for Slimmer Fine-Tuned Models and Better Transfer ACL 2025 NeuroAda: Activating Each Neuron’s Potential for Parameter-Efficient Fine-Tuning EMNLP 2025 NLoRA: Nyström-Initiated Low-Rank Adaptation for Large Language Models EMNLP 2025 Not All Layers of LLMs Are Necessary During Inference IJCAI 2025 Not All Parameters Matter: Masking Diffusion Models for Enhancing Generation Ability CVPR 2025 Numerical Pruning for Efficient Autoregressive Models AAAI 2025 OAC: Output-adaptive Calibration for Accurate Post-training Quantization AAAI 2025 OmniCache: A Trajectory-Oriented Global Perspective on Training-Free Cache Reuse for Diffusion Transformer Models ICCV 2025 One-for-All Pruning: A Universal Model for Customized Compression of Large Language Models ACL 2025 One QuantLLM for ALL: Fine-tuning Quantized LLMs Once for Efficient Deployments ACL 2025 Optimizing Neural Network Training and Quantization with Rooted Logistic Objectives AISTATS 2025 Optimizing Quantized Diffusion Models via Distillation with Cross-Timestep Error Correction AAAI 2025 Optimizing Trajectory Matching Distillation via Parameter Difference-Driven Pruning ACML 2025 ORAL: Prompting Your Large-Scale LoRAs via Conditional Recurrent Diffusion EMNLP 2025 Orthogonal Finetuning Made Scalable EMNLP 2025