Model Compression

MixA-Q: Revisiting Activation Sparsity for Vision Transformers from a Mixed-Precision Quantization Perspective ICCV 2025 Mixture of Weight-shared Heterogeneous Group Attention Experts for Dynamic Token-wise KV Optimization EMNLP 2025 MLAS-LoRA: Language-Aware Parameters Detection and LoRA-Based Knowledge Transfer for Multilingual Machine Translation ACL 2025 MLWQ: Efficient Small Language Model Deployment via Multi-Level Weight Quantization EMNLP 2025 MobiLoRA: Accelerating LoRA-based LLM Inference on Mobile Devices via Context-aware KV Cache Optimization ACL 2025 MOBIUS: Big-to-Mobile Universal Instance Segmentation via Multi-modal Bottleneck Fusion and Calibrated Decoder Pruning ICCV 2025 MobiZO: Enabling Efficient LLM Fine-Tuning at the Edge via Inference Engines EMNLP 2025 Model Unlearning via Sparse Autoencoder Subspace Guided Projections EMNLP 2025 MoLA: MoE LoRA with Layer-wise Expert Allocation NAACL 2025 MoQAE: Mixed-Precision Quantization for Long-Context LLM Inference via Mixture of Quantization-Aware Experts 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 MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation COLING 2025 MT2ST: Adaptive Multi-Task to Single-Task Learning ACL 2025 Multi-Attribute Steering of Language Models via Targeted Intervention ACL 2025 Multilingual Iterative Model Pruning: What Matters? IJCNLP 2025 Multi-Modality Expansion and Retention for LLMs through Parameter Merging and Decoupling ACL 2025 Multi-Programming Language Sandbox for LLMs ACL 2025 MUNBa: Machine Unlearning via Nash Bargaining ICCV 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 NN-Former: Rethinking Graph Structure in Neural Architecture Representation CVPR 2025 Not All Parameters Are Created Equal: Smart Isolation Boosts Fine-Tuning Performance EMNLP 2025