Model Compression

Threshold-driven Pruning with Segmented Maximum Term Weights for Approximate Cluster-based Sparse Retrieval EMNLP 2024 Tied-LoRA: Enhancing parameter efficiency of LoRA with Weight Tying NAACL 2024 To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models EMNLP 2024 Token Alignment via Character Matching for Subword Completion ACL 2024 Towards an On-device Agent for Text Rewriting NAACL 2024 Towards Cheaper Inference in Deep Networks with Lower Bit-Width Accumulators ICLR 2024 Towards efficient deep spiking neural networks construction with spiking activity based pruning ICML 2024 Towards Energy Efficient Spiking Neural Networks: An Unstructured Pruning Framework ICLR 2024 Towards Intelligent Automobile Cockpit via A New Container Architecture NSDI 2024 Towards Next-Level Post-Training Quantization of Hyper-Scale Transformers NeurIPS 2024 Towards Reliable and Efficient Backdoor Trigger Inversion via Decoupling Benign Features ICLR 2024 Training Binary Neural Networks via Gaussian Variational Inference and Low-Rank Semidefinite Programming NeurIPS 2024 Tree Cross Attention ICLR 2024 UltraSparseBERT: 99% Conditionally Sparse Language Modelling ACL 2024 Uncovering the Redundancy in Graph Self-supervised Learning Models NeurIPS 2024 UniPTS: A Unified Framework for Proficient Post-Training Sparsity CVPR 2024 Unlearning Traces the Influential Training Data of Language Models ACL 2024 Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression ACL 2024 Unlocking Memorization in Large Language Models with Dynamic Soft Prompting EMNLP 2024 Unraveling Babel: Exploring Multilingual Activation Patterns of LLMs and Their Applications EMNLP 2024 Unveiling LoRA Intrinsic Ranks via Salience Analysis NeurIPS 2024 USM RNN-T model weights binarization Interspeech 2024 Variation Spaces for Multi-Output Neural Networks: Insights on Multi-Task Learning and Network Compression JMLR 2024 VB-LoRA: Extreme Parameter Efficient Fine-Tuning with Vector Banks NeurIPS 2024 VeRA: Vector-based Random Matrix Adaptation ICLR 2024