Results 241 to 250 of about 114,287 (303)

Prune perineum

Teratology, 1979
AbstractAn infant, horn to unrelated parents, who had a rugated perineal mass which measured 17 cm in diameter is reported. No external genitalia or anal orifice was identified although the infant voided from a 5 mm crevice on the caudal surface of the mass. The patient died at four weeks of age. The perineal mass was made up of two separate sacs.
J N, Peeden, R S, Wilroy, R G, Soper
openaire   +2 more sources

Token Pruning in Multimodal Large Language Models: Are We Solving the Right Problem?

Annual Meeting of the Association for Computational Linguistics
Multimodal large language models (MLLMs) have shown remarkable performance for cross-modal understanding and generation, yet still suffer from severe inference costs.
Zichen Wen, Yifeng Gao, Weijia Li, Conghui He, Linfeng Zhang   +4 more
semanticscholar   +1 more source

Compact Language Models via Pruning and Knowledge Distillation

Neural Information Processing Systems
Large language models (LLMs) targeting different deployment scales and sizes are currently produced by training each variant from scratch; this is extremely compute-intensive.
Saurav Muralidharan, Sharath Turuvekere Sreenivas, Raviraj Joshi, Marcin Chochowski, M. Patwary, M. Shoeybi, Bryan Catanzaro, Jan Kautz, Pavlo Molchanov   +8 more
semanticscholar   +1 more source

Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications

International Conference on Machine Learning
Large language models (LLMs) show inherent brittleness in their safety mechanisms, as evidenced by their susceptibility to jailbreaking and even non-malicious fine-tuning. This study explores this brittleness of safety alignment by leveraging pruning and
Boyi Wei, Kaixuan Huang, Yangsibo Huang, Tinghao Xie, Xiangyu Qi, Mengzhou Xia, Prateek Mittal, Mengdi Wang, Peter Henderson   +8 more
semanticscholar   +1 more source

Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks

International Symposium on Recent Advances in Intrusion Detection, 2018
Deep neural networks (DNNs) provide excellent performance across a wide range of classification tasks, but their training requires high computational resources and is often outsourced to third parties.
Kang Liu, Brendan Dolan-Gavitt, S. Garg
semanticscholar   +1 more source

IRPruneDet: Efficient Infrared Small Target Detection via Wavelet Structure-Regularized Soft Channel Pruning

AAAI Conference on Artificial Intelligence
Infrared Small Target Detection (IRSTD) refers to detecting faint targets in infrared images, which has achieved notable progress with the advent of deep learning.
Mingjin Zhang, Handi Yang, Jie Guo, Yunsong Li, Xinbo Gao, Jing Zhang   +5 more
semanticscholar   +1 more source

TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model

Computer Vision and Pattern Recognition
Vision-Language Models (VLMs) demand substantial computational resources during inference, largely due to the extensive visual input tokens for representing visual information. Previous studies have noted that visual tokens tend to receive less attention
Cheng Yang, Yang Sui, Jinqi Xiao, Lingyi Huang, Yu Gong, Chendi Li, J. Yan, Yu Bai, P. Sadayappan, X. Hu, Bo Yuan   +10 more
semanticscholar   +1 more source

Enhanced forward pruning

Information Sciences, 2005
In this paper forward-pruning methods, such as multi-cut and null move, are tested at so-called ALL nodes. We improved the principal variation search by four small but essential additions. The new PVS algorithm guarantees that forward pruning is safe at ALL nodes.
Mark H. M. Winands, H. Jaap van den Herik, Jos W. H. M. Uiterwijk, Erik C. D. van der Werf   +3 more
openaire   +3 more sources

LaCo: Large Language Model Pruning via Layer Collapse

Conference on Empirical Methods in Natural Language Processing
Large language models (LLMs) based on transformer are witnessing a notable trend of size expansion, which brings considerable costs to both model training and inference.
Yifei Yang, Zouying Cao, Hai Zhao
semanticscholar   +1 more source

Pruning

2000
Abstract The pruning of trees is probably the most noticeable and important of all tree maintenance practices. Thoughtful pruning produces a structurally sound tree that can better withstand adverse environmental conditions. In addition, properly pruned trees require less cabling, bracing, and sometimes require less managing of pests to ...
John R Hartman, Thomas P Pirone, Mary Ann Sall   +2 more
  +4 more sources