Results 21 to 30 of about 4,189,347 (339)
Editing Large Language Models: Problems, Methods, and Opportunities [PDF]
Conference on Empirical Methods in Natural Language Processing, 2023 Despite the ability to train capable LLMs, the methodology for maintaining their relevancy and rectifying errors remains elusive. To this end, the past few years have witnessed a surge in techniques for editing LLMs, the objective of which is to ...
Yunzhi Yao +7 more
semanticscholar +1 more source
Editing Models with Task Arithmetic [PDF]
International Conference on Learning Representations, 2022 Changing how pre-trained models behave -- e.g., improving their performance on a downstream task or mitigating biases learned during pre-training -- is a common practice when developing machine learning systems.
Gabriel Ilharco +6 more
semanticscholar +1 more source
Mass-Editing Memory in a Transformer [PDF]
International Conference on Learning Representations, 2022 Recent work has shown exciting promise in updating large language models with new memories, so as to replace obsolete information or add specialized knowledge. However, this line of work is predominantly limited to updating single associations.
Kevin Meng +4 more
semanticscholar +1 more source
Emu Edit: Precise Image Editing via Recognition and Generation Tasks [PDF]
Computer Vision and Pattern Recognition, 2023 Instruction-based image editing holds immense potential for a variety of applications, as it enables users to perform any editing operation using a natural language instruction.
Shelly Sheynin +7 more
semanticscholar +1 more source
Pix2Video: Video Editing using Image Diffusion [PDF]
IEEE International Conference on Computer Vision, 2023 Image diffusion models, trained on massive image collections, have emerged as the most versatile image generator model in terms of quality and diversity.
Duygu Ceylan, C. Huang, N. Mitra
semanticscholar +1 more source
Rewriting Human History and Empowering Indigenous Communities with Genome Editing Tools. [PDF]
, 2020 Appropriate empirical-based evidence and detailed theoretical considerations should be used for evolutionary explanations of phenotypic variation observed in the field of human population genetics (especially Indigenous populations). Investigators within
Fox, Keolu +2 more
core +2 more sources
Blended Diffusion for Text-driven Editing of Natural Images [PDF]
Computer Vision and Pattern Recognition, 2021 Natural language offers a highly intuitive interface for image editing. In this paper, we introduce the first solution for performing local (region-based) edits in generic natural images, based on a natural language description along with an ROI mask. We
Omri Avrahami, D. Lischinski, Ohad Fried
semanticscholar +1 more source
Deep sampling of gRNA in the human genome and deep-learning-informed prediction of gRNA activities
Cell Discovery, 2023 Life science studies involving clustered regularly interspaced short palindromic repeat (CRISPR) editing generally apply the best-performing guide RNA (gRNA) for a gene of interest.
Heng Zhang +8 more
doaj +1 more source
FOSP’S “Publishing Passion”: on the Materials Held in the Department of Manuscripts of the IWL RAS [PDF]
Studia Litterarum, 2020 This article based on the archival material from the Department of manuscripts of IWL RAN discusses the hitherto understudied financial and ideological conditions of the Publishing house of the Federation of Associations of Soviet writers (FOSP ...
Daria S. Moskovskaya, Elena D. Galtsova
doaj +1 more source
The necessity for reediting The Nozhatnāmeh ‘Alā’i [PDF]
تاریخ ادبیات, 2021 Nozhatnāmeh ‘Alā’i by Shahmardan ibn Abi al-Khair Razi, az one of the first Persian encyclopedia, belongs to the most important class of prose writings.
Masumeh Samkhaniani +3 more
doaj +1 more source