Results 11 to 20 of about 13,789,012 (283)
Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluation [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Few-shot fine-tuning and in-context learning are two alternative strategies for task adaptation of pre-trained language models. Recently, in-context learning has gained popularity over fine-tuning due to its simplicity and improved out-of-domain ...
Marius Mosbach +4 more
semanticscholar +1 more source
What Makes Good Examples for Visual In-Context Learning? [PDF]
Neural Information Processing Systems, 2023 Large-scale models trained on broad data have recently become the mainstream architecture in computer vision due to their strong generalization performance.
Yuanhan Zhang, Kaiyang Zhou, Ziwei Liu
semanticscholar +1 more source
The Learnability of In-Context Learning [PDF]
Neural Information Processing Systems, 2023 In-context learning is a surprising and important phenomenon that emerged when modern language models were scaled to billions of learned parameters. Without modifying a large language model's weights, it can be tuned to perform various downstream natural
Noam Wies, Yoav Levine, A. Shashua
semanticscholar +1 more source
Barometric Trend in Contemporary Design A Study in Bricks Facades [PDF]
Engineering and Technology Journal, 2017 Recently appeared many trends in architecture design, one of theme is “Barometric Design”, especially after the emergence of digital-oriented modeling programs within Construction, which needs high accuracy, and Craft.
A.M.H. Al-Moqaram, S. R.H. Majed
doaj +1 more source
EAI Endorsed Transactions on Industrial Networks and Intelligent Systems, 2023 With the increasing prevalence of the Internet of Things (IoT), there is a growing need for effective access control methods to secure IoT systems and data.
Priyanka More, Sachin Sakhare
doaj +1 more source
QUALITY PARAMETERS OF INFORMATION SYSTEMS FOR BUSINESS IN THE CONTEXT OF DIGITAL TRANSFORMATIONS [PDF]
E3S Web of Conferences, 2020 The article provides a refined definition of “information system for business” as a coordinated set of material, non-material and human re-sourses components that is used to implement a set of procedures of form¬ing an information resource as a quality ...
Abu Ezza Hasan +3 more
doaj +1 more source
In-context Autoencoder for Context Compression in a Large Language Model [PDF]
International Conference on Learning Representations, 2023 We propose the In-context Autoencoder (ICAE), leveraging the power of a large language model (LLM) to compress a long context into short compact memory slots that can be directly conditioned on by the LLM for various purposes.
Tao Ge +4 more
semanticscholar +1 more source
Are Emergent Abilities in Large Language Models just In-Context Learning? [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Large language models, comprising billions of parameters and pre-trained on extensive web-scale corpora, have been claimed to acquire certain capabilities without having been specifically trained on them.
Sheng Lu +4 more
semanticscholar +1 more source
Backdoor Attacks for In-Context Learning with Language Models [PDF]
arXiv.org, 2023 Because state-of-the-art language models are expensive to train, most practitioners must make use of one of the few publicly available language models or language model APIs. This consolidation of trust increases the potency of backdoor attacks, where an
Nikhil Kandpal +3 more
semanticscholar +1 more source
Urban, Planning and Transport Research, 2023 Road accidents generate substantial personal and national losses. Road accidents reduce victims' and their families' productivity, which slows the nation's economic growth. The victim or victim's heir must be compensated for the mishap's losses.
Adil Ata Azmi, Sewa Ram
doaj +1 more source