Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Moskvichev, Arseny"'
Publikováno v:
Proceedings of the LLM-CP Workshop, AAAI 2024
We explore the abstract reasoning abilities of text-only and multimodal versions of GPT-4, using the ConceptARC benchmark [10], which is designed to evaluate robust understanding and reasoning with core-knowledge concepts. We extend the work of Moskv
Externí odkaz:
http://arxiv.org/abs/2311.09247
Autor:
Moskvichev, Arseny, Mai, Ky-Vinh
We propose a new large-scale (nearly a million questions) ultra-long-context (more than 50,000 words average document length) reading comprehension dataset. Using GPT 3.5, we summarized each scene in 1,500 hand-curated fiction books from Project Gute
Externí odkaz:
http://arxiv.org/abs/2305.13877
Publikováno v:
Transactions on Machine Learning Research, 8/2023
The abilities to form and abstract concepts is key to human intelligence, but such abilities remain lacking in state-of-the-art AI systems. There has been substantial research on conceptual abstraction in AI, particularly using idealized domains such
Externí odkaz:
http://arxiv.org/abs/2305.07141
Autor:
Moskvichev, Arseny, Liu, James A.
Developing NLP models traditionally involves two stages - training and application. Retention of information acquired after training (at application time) is architecturally limited by the size of the model's context window (in the case of transforme
Externí odkaz:
http://arxiv.org/abs/2104.05500
Publikováno v:
1st Workshop on Language in Reinforcement Learning, ICML 2020
Social network structure is one of the key determinants of human language evolution. Previous work has shown that the network of social interactions shapes decentralized learning in human groups, leading to the emergence of different kinds of communi
Externí odkaz:
http://arxiv.org/abs/2007.09820
Publikováno v:
In Cognition September 2023 238
Autor:
Moskvichev, Arseny, Mai, Ky-Vinh
Despite their tremendous successes, most large language models do not have any long-term memory mechanisms, which restricts their applications. Overcoming this limitation would not only require changes to the typical transformer architectures or trai
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4271f8484def112268303512f2672550
http://arxiv.org/abs/2305.13877
http://arxiv.org/abs/2305.13877
Autor:
Dubova, Marina, Moskvichev, Arseny
This project aims to study the role of active perceptual strategies and nameability in learning relations between objects (e.g. A is same/different to B, A is/isn't part of B, etc.). We propose a range of tasks where participants need to iteratively
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a572093dc57db1d37a1dda7670101188