v1v2v3 (latest)

Editing Models with Task Arithmetic

International Conference on Learning Representations (ICLR), 2022

8 December 2022

ArXiv (abs)PDF HTML HuggingFace (7 upvotes)

Papers citing "Editing Models with Task Arithmetic"

50 / 523 papers shown

Title
A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and ToxicityInternational Conference on Machine Learning (ICML), 2024 Andrew Lee Xiaoyan Bai Itamar Pres Martin Wattenberg Jonathan K. Kummerfeld Amélie Reymond 282 157 0 03 Jan 2024
A Comprehensive Study of Knowledge Editing for Large Language Models Ningyu Zhang Yunzhi Yao Bo Tian Peng Wang Shumin Deng ... Lei Liang Qing Cui Xiao-Jun Zhu Jun Zhou Huajun Chen KELM 445 124 0 02 Jan 2024
Partial Fine-Tuning: A Successor to Full Fine-Tuning for Vision Transformers Peng Ye Yongqi Huang Chongjun Tu Minglei Li Tao Chen Tong He Wanli Ouyang 177 14 0 25 Dec 2023
Merging Vision Transformers from Different Tasks and Domains Peng Ye Chenyu Huang Mingzhu Shen Tao Chen Yongqi Huang Yuning Zhang Wanli Ouyang MoMe 197 16 0 25 Dec 2023
Multimodal Attention Merging for Improved Speech Recognition and Audio Event Classification Anirudh S. Sundar Chao-Han Huck Yang David M. Chan Shalini Ghosh Venkatesh Ravichandran P. S. Nidadavolu MoMe 246 12 0 22 Dec 2023
Parameter-Efficient Fine-Tuning Methods for Pretrained Language Models: A Critical Review and Assessment Lingling Xu Haoran Xie S. J. Qin Xiaohui Tao F. Wang 265 261 0 19 Dec 2023
Model Breadcrumbs: Scaling Multi-Task Model Merging with Sparse Masks Mohammad-Javad Davari Eugene Belilovsky MoMe 244 96 0 11 Dec 2023
Concrete Subspace Learning based Interference Elimination for Multi-task Model Fusion Anke Tang Li Shen Yong Luo Liang Ding Han Hu Bo Du Dacheng Tao MoMe 270 30 0 11 Dec 2023
Merging by Matching Models in Task Parameter Subspaces Derek Tam Mohit Bansal Colin Raffel MoMe 317 20 0 07 Dec 2023
Knowledge Unlearning for LLMs: Tasks, Methods, and Challenges Nianwen Si Hao Zhang Heyu Chang Wenlin Zhang Dan Qu Weiqiang Zhang KELM MU 374 38 0 27 Nov 2023
ComPEFT: Compression for Communicating Parameter Efficient Updates via Sparsification and Quantization Prateek Yadav Leshem Choshen Colin Raffel Mohit Bansal 234 18 0 22 Nov 2023
In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space SteeringInternational Conference on Machine Learning (ICML), 2023 Sheng Liu Haotian Ye Lei Xing James Y. Zou 238 199 0 11 Nov 2023
LCM-LoRA: A Universal Stable-Diffusion Acceleration Module Simian Luo Yiqin Tan Suraj Patil Daniel Gu Patrick von Platen Apolinário Passos Longbo Huang Jian Li Hang Zhao MoMe 574 204 0 09 Nov 2023
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free LunchInternational Conference on Machine Learning (ICML), 2023 Le Yu Yu Bowen Haiyang Yu Fei Huang Yongbin Li MoMe 498 480 0 06 Nov 2023
A Survey on Knowledge Editing of Neural NetworksIEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2023 Vittorio Mazzia Alessandro Pedrani Andrea Caciolai Kay Rottmann Davide Bernardi KELM 361 37 0 30 Oct 2023
SoK: Memorization in General-Purpose Large Language Models Valentin Hartmann Anshuman Suri Vincent Bindschaedler David Evans Shruti Tople Robert West KELM LLMAG 304 36 0 24 Oct 2023
SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding Haoxiang Wang Pavan Kumar Anasosalu Vasu Fartash Faghri Raviteja Vemulapalli Mehrdad Farajtabar Sachin Mehta Mohammad Rastegari Oncel Tuzel Hadi Pouransari VLM 536 123 0 23 Oct 2023
Function Vectors in Large Language ModelsInternational Conference on Learning Representations (ICLR), 2023 Eric Todd Millicent Li Arnab Sen Sharma Aaron Mueller Byron C. Wallace David Bau 287 181 0 23 Oct 2023
Equivariant Deep Weight Space Alignment Aviv Navon Aviv Shamsian Ethan Fetaya Gal Chechik Nadav Dym Haggai Maron 355 29 0 20 Oct 2023
Model Merging by Uncertainty-Based Gradient Matching Nico Daheim Thomas Möllenhoff Edoardo Ponti Iryna Gurevych Mohammad Emtiyaz Khan MoMe FedML 293 72 0 19 Oct 2023
Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging Joel Jang Seungone Kim Bill Yuchen Lin Yizhong Wang Jack Hessel Luke Zettlemoyer Hannaneh Hajishirzi Yejin Choi Prithviraj Ammanabrolu MoMe 309 210 0 17 Oct 2023
Seeking Neural Nuggets: Knowledge Transfer in Large Language Models from a Parametric PerspectiveInternational Conference on Learning Representations (ICLR), 2023 Ming Zhong Chenxin An Weizhu Chen Jiawei Han Pengcheng He 340 16 0 17 Oct 2023
Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingInternational Conference on Learning Representations (ICLR), 2023 Melanie Sclar Yejin Choi Yulia Tsvetkov Alane Suhr 293 532 0 17 Oct 2023
Can We Edit Multimodal Large Language Models?Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023 Siyuan Cheng Bo Tian Qingbin Liu Xi Chen Yongheng Wang Huajun Chen Ningyu Zhang MLLM 539 38 0 12 Oct 2023
Measuring Feature Sparsity in Language Models Mingyang Deng Lucas Tao Joe Benton 214 2 0 11 Oct 2023
A Meta-Learning Perspective on Transformers for Causal Language ModelingAnnual Meeting of the Association for Computational Linguistics (ACL), 2023 Xinbo Wu Lav Varshney 273 8 0 09 Oct 2023
Establishing Trustworthiness: Rethinking Tasks and Model EvaluationConference on Empirical Methods in Natural Language Processing (EMNLP), 2023 Robert Litschko Max Müller-Eberstein Rob van der Goot Leon Weber Barbara Plank LRM 170 3 0 09 Oct 2023
Uncovering hidden geometry in Transformers via disentangling position and context Jiajun Song Yiqiao Zhong 224 13 0 07 Oct 2023
Parameter Efficient Multi-task Model Fusion with Partial LinearizationInternational Conference on Learning Representations (ICLR), 2023 Anke Tang Li Shen Yong Luo Yibing Zhan Han Hu Bo Du Yixin Chen Dacheng Tao MoMe 319 52 0 07 Oct 2023
AdaMerging: Adaptive Model Merging for Multi-Task LearningInternational Conference on Learning Representations (ICLR), 2023 Enneng Yang Zhenyi Wang Li Shen Shiwei Liu Guibing Guo Xingwei Wang Dacheng Tao MoMe 300 180 0 04 Oct 2023
BYOM: Building Your Own Multi-Task Model For Free Weisen Jiang Xiaoyuan Zhang Han Shi Yu Zhang Zhenguo Li James T. Kwok MoMe 254 6 0 03 Oct 2023
Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing PolicyInternational Conference on Learning Representations (ICLR), 2023 Pingzhi Li Zhenyu Zhang Prateek Yadav Yi-Lin Sung Yu Cheng Mohit Bansal Tianlong Chen MoMe 245 73 0 02 Oct 2023
ScaLearn: Simple and Highly Parameter-Efficient Task Transfer by Learning to ScaleAnnual Meeting of the Association for Computational Linguistics (ACL), 2023 Markus Frohmann Carolin Holtermann Shahed Masoudian Anne Lauscher Navid Rekabsaz 325 2 0 02 Oct 2023
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction AttacksInternational Conference on Learning Representations (ICLR), 2023 Vaidehi Patil Peter Hase Joey Tianyi Zhou KELM AAML 289 144 0 29 Sep 2023
Deep Model Fusion: A Survey Weishi Li Yong Peng Miao Zhang Liang Ding Han Hu Li Shen FedML MoMe 277 86 0 27 Sep 2023
Knowledge Sanitization of Large Language Models Yoichi Ishibashi Hidetoshi Shimodaira KELM 231 35 0 21 Sep 2023
Cognitive Mirage: A Review of Hallucinations in Large Language Models Hongbin Ye Tong Liu Aijia Zhang Wei Hua Weiqiang Jia HILM 357 111 0 13 Sep 2023
Circuit Breaking: Removing Model Behaviors with Targeted Ablation Maximilian Li Xander Davies Max Nadeau KELM MU 275 34 0 12 Sep 2023
Emergent Linear Representations in World Models of Self-Supervised Sequence ModelsBlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP), 2023 Neel Nanda Andrew Lee Martin Wattenberg FAtt MILM 279 246 0 02 Sep 2023
Fine-tuning can cripple your foundation model; preserving features may be the solution Jishnu Mukhoti Y. Gal Juil Sock P. Dokania CLL 361 67 0 25 Aug 2023
Overcoming Generic Knowledge Loss with Selective Parameter UpdateComputer Vision and Pattern Recognition (CVPR), 2023 Wenxuan Zhang Paul Janson Rahaf Aljundi Mohamed Elhoseiny KELM CLL 339 19 0 23 Aug 2023
UnIVAL: Unified Model for Image, Video, Audio and Language Tasks Mustafa Shukor Corentin Dancette Alexandre Ramé Matthieu Cord MoMe MLLM 271 54 0 30 Jul 2023
LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Composition Chengsong Huang Qian Liu Bill Yuchen Lin Tianyu Pang Chao Du Min Lin MoMe 433 289 0 25 Jul 2023
Layer-wise Linear Mode ConnectivityInternational Conference on Learning Representations (ICLR), 2023 Linara Adilova Maksym Andriushchenko Michael Kamp Asja Fischer Martin Jaggi FedML FAtt MoMe 457 20 0 13 Jul 2023
STG-MTL: Scalable Task Grouping for Multi-Task Learning Using Data Map Ammar Sherif Abubakar Abid M. Elattar Mohamed ElHelw 385 6 0 07 Jul 2023
ProbVLM: Probabilistic Adapter for Frozen Vision-Language Models Uddeshya Upadhyay Shyamgopal Karthik Goran Frehse Zeynep Akata MLLM VLM 434 5 0 01 Jul 2023
Composing Parameter-Efficient Modules with Arithmetic OperationsNeural Information Processing Systems (NeurIPS), 2023 Jinghan Zhang Shiqi Chen Junteng Liu Junxian He KELM MoMe 335 152 0 26 Jun 2023
Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewardsNeural Information Processing Systems (NeurIPS), 2023 Alexandre Ramé Guillaume Couairon Mustafa Shukor Corentin Dancette Jean-Baptiste Gaya Laure Soulier Matthieu Cord MoMe 347 201 0 07 Jun 2023
Inference-Time Intervention: Eliciting Truthful Answers from a Language ModelNeural Information Processing Systems (NeurIPS), 2023 Kenneth Li Oam Patel Fernanda Viégas Hanspeter Pfister Martin Wattenberg KELM HILM 699 809 0 06 Jun 2023
TIES-Merging: Resolving Interference When Merging ModelsNeural Information Processing Systems (NeurIPS), 2023 Prateek Yadav Derek Tam Leshem Choshen Colin Raffel Joey Tianyi Zhou MoMe 378 506 0 02 Jun 2023