How to think step-by-step: A mechanistic understanding of chain-of-thought reasoningby Subhabrata Dutta, Joykirat Singh, Soumen…
Asymmetry in Low-Rank Adapters of Foundation Modelsby Jiacheng Zhu, Kristjan Greenewald, Kimia Nadjahi, Haitz Sáez…
Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Promptsby Mikayel Samvelyan, Sharath Chandra Raparthy, Andrei Lupu,…
AutoMMLab: Automatically Generating Deployable Models from Language Instructions for Computer Vision Tasksby Zekang Yang, Wang…
GPTVQ: The Blessing of Dimensionality for LLM Quantizationby Mart van Baalen, Andrey Kuzmin, Markus Nagel,…
Advancing Parameter Efficiency in Fine-tuning via Representation Editingby Muling Wu, Wenhao Liu, Xiaohua Wang, Tianlong…
APTQ: Attention-aware Post-Training Mixed-Precision Quantization for Large Language Modelsby Ziyi Guan, Hantao Huang, Yupeng Su,…
Recursive Speculative Decoding: Accelerating LLM Inference via Sampling Without Replacementby Wonseok Jeon, Mukul Gagrani, Raghavv…
Transformer tricks: Precomputing the first layerby Nils GraefFirst submitted to arxiv on: 20 Feb 2024CategoriesMain:…
SPML: A DSL for Defending Language Models Against Prompt Attacksby Reshabh K Sharma, Vinayak Gupta,…