Model Compression

TaDA: Training-free recipe for Decoding with Adaptive KV Cache Compression and Mean-centering ACL 2025 TAID: Temporally Adaptive Interpolated Distillation for Efficient Knowledge Transfer in Language Models ICLR 2025 TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization ACL 2025 Talking Head Anime 4: Distillation for Real-Time Performance WACV 2025 Tamper-Resistant Safeguards for Open-Weight LLMs ICLR 2025 TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Splatting CVPR 2025 Targeted Low-rank Refinement: Enhancing Sparse Language Models with Precision ICML 2025 Task-Specific Zero-shot Quantization-Aware Training for Object Detection ICCV 2025 TASO: Task-Aligned Sparse Optimization for Parameter-Efficient Model Adaptation EMNLP 2025 TBA at BEA 2025 Shared Task: Transfer-Learning from DARE-TIES Merged Models for the Pedagogical Ability Assessment of LLM-Powered Math Tutors ACL 2025 The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws ICLR 2025 The Role of Outgoing Connection Heterogeneity in Feedforward Layers of Large Language Models EMNLP 2025 The Unreasonable Ineffectiveness of the Deeper Layers ICLR 2025 ThinK: Thinner Key Cache by Query-Driven Pruning ICLR 2025 Tiny Budgets, Big Gains: Parameter Placement Strategy in Parameter Super-Efficient Fine-Tuning EMNLP 2025 Token Level Routing Inference System for Edge Devices ACL 2025 TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection EMNLP 2025 TokenSkip: Controllable Chain-of-Thought Compression in LLMs EMNLP 2025 Token Weighting for Long-Range Language Modeling NAACL 2025 Toward Adaptive Large Language Models Structured Pruning via Hybrid-grained Weight Importance Assessment AAAI 2025 Towards Efficient CoT Distillation: Self-Guided Rationale Selector for Better Performance with Fewer Rationales EMNLP 2025 Towards More Efficient Post-training via Fourier Domain Adapter Framework EMNLP 2025 Towards One-bit ASR: Extremely Low-bit Conformer Quantization Using Co-training and Stochastic Precision Interspeech 2025 Training-Free Activation Sparsity in Large Language Models ICLR 2025 Training Free Exponential Context Extension via Cascading KV Cache ICLR 2025