Model Compression

Improving GNN Calibration with Discriminative Ability: Insights and Strategies AAAI 2024 Incremental pre-training from smaller language models ACL 2024 Induced Model Matching: Restricted Models Help Train Full-Featured Models NeurIPS 2024 Initializing Models with Larger Ones ICLR 2024 Initializing Variable-sized Vision Transformers from Learngene with Learnable Transformation NeurIPS 2024 Instance-Aware Group Quantization for Vision Transformers CVPR 2024 IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens Intact ACL 2024 Integer Is Enough: When Vertical Federated Learning Meets Rounding AAAI 2024 Interpretable Mesomorphic Networks for Tabular Data NeurIPS 2024 Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis EMNLP 2024 Is C4 Dataset Optimal for Pruning? An Investigation of Calibration Data for LLM Pruning EMNLP 2024 Is Modularity Transferable? A Case Study through the Lens of Knowledge Distillation COLING 2024 IT-Tuning : Parameter Efficient Information Token Tuning for Language Model ACL 2024 Jointly Training and Pruning CNNs via Learnable Agent Guidance and Alignment CVPR 2024 JointSQ: Joint Sparsification-Quantization for Distributed Learning CVPR 2024 KOALA: Empirical Lessons Toward Memory-Efficient and Fast Diffusion Models for Text-to-Image Synthesis NeurIPS 2024 KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization NeurIPS 2024 LaMDA: Large Model Fine-Tuning via Spectrally Decomposed Low-Dimensional Adaptation EMNLP 2024 Language Models as Zero-shot Lossless Gradient Compressors: Towards General Neural Parameter Prior Models NeurIPS 2024 Layer-Adaptive State Pruning for Deep State Space Models NeurIPS 2024 Layer Compression of Deep Networks with Straight Flows AAAI 2024 LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding ACL 2024 Learning Cut Generating Functions for Integer Programming NeurIPS 2024 Learning Global Controller in Latent Space for Parameter-Efficient Fine-Tuning ACL 2024 Learning Performance Maximizing Ensembles with Explainability Guarantees AAAI 2024