Artificial Intelligence › Core AI ›

Model Compression

1928 directly classified papers

Papers per year

Papers

Leveraging Attention to Effectively Compress Prompts for Long-Context LLMs AAAI 2025

MeRino: Entropy-Driven Design for Generative Language Models on IoT Devices AAAI 2025

QJL: 1-Bit Quantized JL Transform for KV Cache Quantization with Zero Overhead AAAI 2025

PrivDNFIS: Privacy-preserving and Efficient Deep Neuro-Fuzzy Inference System AAAI 2025

Advancing Weight and Channel Sparsification with Enhanced Saliency WACV 2025

CSR:Achieving 1 Bit Key-Value Cache via Sparse Representation AAAI 2025

PAT: Pruning-Aware Tuning for Large Language Models AAAI 2025

When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoning EMNLP 2025

FlexiDiT: Your Diffusion Transformer Can Easily Generate High-Quality Samples with Less Compute CVPR 2025

Not All Layers of LLMs Are Necessary During Inference IJCAI 2025

Reference-Based Post-OCR Processing with LLM for Precise Diacritic Text in Historical Document Recognition AAAI 2025

SparseMM: Head Sparsity Emerges from Visual Concept Responses in MLLMs ICCV 2025

Pioneering 4-Bit FP Quantization for Diffusion Models: Mixup-Sign Quantization and Timestep-Aware Fine-Tuning CVPR 2025

Minimal Interaction Seperated Tuning: A New Paradigm for Visual Adaptation CVPR 2025

TRNAS: A Training-Free Robust Neural Architecture Search ICCV 2025

EdgeTAM: On-Device Track Anything Model CVPR 2025

DeepCompress-ViT: Rethinking Model Compression to Enhance Efficiency of Vision Transformers at the Edge CVPR 2025

Efficient On-Device Text Simplification for Firefox with Synthetic Data Fine-Tuning EMNLP 2025

OneNRC@TSAR2025 Shared Task Small Models for Readability Controlled Text Simplification EMNLP 2025

Findings of the WMT 2025 Shared Task on Model Compression: Early Insights on Compressing LLMs for Machine Translation EMNLP 2025

Submission for WMT25 Task 3 EMNLP 2025

Iterative Layer Pruning for Efficient Translation Inference EMNLP 2025

MoDification: Mixture of Depths Made Easy NAACL 2025

Vicomtech@WMT 2025: Evolutionary Model Compression for Machine Translation EMNLP 2025

Efficient One-shot Compression via Low-Rank Local Feature Distillation NAACL 2025