Zobrazeno 1 - 10
of 12 094
pro vyhledávání: '"A. Nakov"'
Autor:
Sánchez-Baracaldo, Patricia, Bianchini, Giorgio, Huelsenbeck, John P., Raven, John A., Pisani, Davide, Knoll, Andrew H.
Publikováno v:
Proceedings of the National Academy of Sciences of the United States of America, 2017 Dec 01. 114(50), E10608-E10609.
Externí odkaz:
https://www.jstor.org/stable/26485810
Autor:
Li, Haonan, Han, Xudong, Zhai, Zenan, Mu, Honglin, Wang, Hao, Zhang, Zhenxuan, Geng, Yilin, Lin, Shom, Wang, Renxi, Shelmanov, Artem, Qi, Xiangyu, Wang, Yuxia, Hong, Donghai, Yuan, Youliang, Chen, Meng, Tu, Haoqin, Koto, Fajri, Kuribayashi, Tatsuki, Zeng, Cong, Bhardwaj, Rishabh, Zhao, Bingchen, Duan, Yawen, Liu, Yi, Alghamdi, Emad A., Yang, Yaodong, Dong, Yinpeng, Poria, Soujanya, Liu, Pengfei, Liu, Zhengzhong, Ren, Xuguang, Hovy, Eduard, Gurevych, Iryna, Nakov, Preslav, Choudhury, Monojit, Baldwin, Timothy
To address this gap, we introduce Libra-Leaderboard, a comprehensive framework designed to rank LLMs through a balanced evaluation of performance and safety. Combining a dynamic leaderboard with an interactive LLM arena, Libra-Leaderboard encourages
Externí odkaz:
http://arxiv.org/abs/2412.18551
Jailbreaking in Large Language Models (LLMs) is a major security concern as it can deceive LLMs to generate harmful text. Yet, there is still insufficient understanding of how jailbreaking works, which makes it hard to develop effective defense strat
Externí odkaz:
http://arxiv.org/abs/2412.17034
Autor:
Xu, Jundong, Fei, Hao, Luo, Meng, Liu, Qian, Pan, Liangming, Wang, William Yang, Nakov, Preslav, Lee, Mong-Li, Hsu, Wynne
In the context of large language models (LLMs), current advanced reasoning methods have made impressive strides in various reasoning tasks. However, when it comes to logical reasoning tasks, major challenges remain in both efficacy and efficiency. Th
Externí odkaz:
http://arxiv.org/abs/2412.16953
Retrieval-Augmented Generation (RAG) systems have emerged as a promising solution to mitigate LLM hallucinations and enhance their performance in knowledge-intensive domains. However, these systems are vulnerable to adversarial poisoning attacks, whe
Externí odkaz:
http://arxiv.org/abs/2412.16708
In the current era of rapidly growing digital data, evaluating the political bias and factuality of news outlets has become more important for seeking reliable information online. In this work, we study the classification problem of profiling news me
Externí odkaz:
http://arxiv.org/abs/2412.10467
Large Language Models (LLMs) have demonstrated remarkable planning abilities across various domains, including robotics manipulation and navigation. While recent efforts in robotics have leveraged LLMs both for high-level and low-level planning, thes
Externí odkaz:
http://arxiv.org/abs/2411.17636
Autor:
Ma, Congbo, Wang, Hu, Qiu, Zitai, Xue, Shan, Wu, Jia, Yang, Jian, Nakov, Preslav, Sheng, Quan Z.
Social media data is inherently rich, as it includes not only text content, but also users, geolocation, entities, temporal information, and their relationships. This data richness can be effectively modeled using heterogeneous information networks (
Externí odkaz:
http://arxiv.org/abs/2411.12588
Publikováno v:
TIP Revista Especializada en Ciencias Químico-Biológicas. 2021, Vol. 24, p1-10. 10p.
The growing use of large language models (LLMs) has raised concerns regarding their safety. While many studies have focused on English, the safety of LLMs in Arabic, with its linguistic and cultural complexities, remains under-explored. Here, we aim
Externí odkaz:
http://arxiv.org/abs/2410.17040