Zobrazeno 1 - 10
of 4 760
pro vyhledávání: '"Tsvetkov, P."'
Autor:
Sclar, Melanie, Yu, Jane, Fazel-Zarandi, Maryam, Tsvetkov, Yulia, Bisk, Yonatan, Choi, Yejin, Celikyilmaz, Asli
Do large language models (LLMs) have theory of mind? A plethora of papers and benchmarks have been introduced to evaluate if current models have been able to develop this key ability of social intelligence. However, all rely on limited datasets with
Externí odkaz:
http://arxiv.org/abs/2412.12175
Conventional algorithms for training language models (LMs) with human feedback rely on preferences that are assumed to account for an "average" user, disregarding subjectivity and finer-grained variations. Recent studies have raised concerns that agg
Externí odkaz:
http://arxiv.org/abs/2410.16027
Autor:
Tsvetkov, Petr, Eliseeva, Aleksandra, Dig, Danny, Bezzubov, Alexander, Golubev, Yaroslav, Bryksin, Timofey, Zharov, Yaroslav
Commit message generation (CMG) is a crucial task in software engineering that is challenging to evaluate correctly. When a CMG system is integrated into the IDEs and other products at JetBrains, we perform online evaluation based on user acceptance
Externí odkaz:
http://arxiv.org/abs/2410.12046
Autor:
Feng, Shangbin, Wang, Zifeng, Wang, Yike, Ebrahimi, Sayna, Palangi, Hamid, Miculicich, Lesly, Kulshrestha, Achin, Rauschmayr, Nathalie, Choi, Yejin, Tsvetkov, Yulia, Lee, Chen-Yu, Pfister, Tomas
We propose Model Swarms, a collaborative search algorithm to adapt LLMs via swarm intelligence, the collective behavior guiding individual systems. Specifically, Model Swarms starts with a pool of LLM experts and a utility function. Guided by the bes
Externí odkaz:
http://arxiv.org/abs/2410.11163
In the absence of abundant reliable annotations for challenging tasks and contexts, how can we expand the frontier of LLM capabilities with potentially wrong answers? We focus on two research questions: (1) Can LLMs generate reliable preferences amon
Externí odkaz:
http://arxiv.org/abs/2410.11055
Autor:
Fisher, Jillian, Feng, Shangbin, Aron, Robert, Richardson, Thomas, Choi, Yejin, Fisher, Daniel W., Pan, Jennifer, Tsvetkov, Yulia, Reinecke, Katharina
As modern AI models become integral to everyday tasks, concerns about their inherent biases and their potential impact on human decision-making have emerged. While bias in models are well-documented, less is known about how these biases influence hum
Externí odkaz:
http://arxiv.org/abs/2410.06415
To explain social phenomena and identify systematic biases, much research in computational social science focuses on comparative text analyses. These studies often rely on coarse corpus-level statistics or local word-level analyses, mainly in English
Externí odkaz:
http://arxiv.org/abs/2410.04282
Autor:
Chiu, Yu Ying, Jiang, Liwei, Lin, Bill Yuchen, Park, Chan Young, Li, Shuyue Stella, Ravi, Sahithya, Bhatia, Mehar, Antoniak, Maria, Tsvetkov, Yulia, Shwartz, Vered, Choi, Yejin
To make large language models (LLMs) more helpful across diverse cultures, it is essential to have effective cultural knowledge benchmarks to measure and track our progress. Effective benchmarks need to be robust, diverse, and challenging. We introdu
Externí odkaz:
http://arxiv.org/abs/2410.02677
Autor:
Azarevich, A. N., Khrykina, O. N., Bolotina, N. B., Gridchina, V. G., Bogach, A. V., Demishev, S. V., Krasnorussky, V. N., Gavrilkin, S. Yu., Tsvetkov, A. Yu., Shitsevalova, N. Yu., Voronov, V. V., Kugel, K. I., Rakhmanov, A. L., Gabani, S., Flachbart, K., Sluchanko, N. E.
The presented studies of resistivity (R), thermal conductivity (k) and specific heat (C) at low temperature 1.8-7 K in magnetic field up to 90 kOe made it possible to detect for the first time the exponential field dependences R(H), 1/k(H), $C(H) \si
Externí odkaz:
http://arxiv.org/abs/2409.04139
Recent work in image and video generation has been adopting the autoregressive LLM architecture due to its generality and potentially easy integration into multi-modal systems. The crux of applying autoregressive training in language generation to vi
Externí odkaz:
http://arxiv.org/abs/2408.08459