Zobrazeno 1 - 10
of 1 268
pro vyhledávání: '"A. Benajiba"'
Autor:
Feng, Yu, Htut, Phu Mon, Qi, Zheng, Xiao, Wei, Mager, Manuel, Pappas, Nikolaos, Halder, Kishaloy, Li, Yang, Benajiba, Yassine, Roth, Dan
Quantifying the uncertainty in the factual parametric knowledge of Large Language Models (LLMs), especially in a black-box setting, poses a significant challenge. Existing methods, which gauge a model's uncertainty through evaluating self-consistency
Externí odkaz:
http://arxiv.org/abs/2412.09572
Autor:
Shahriar, Sadat, Qi, Zheng, Pappas, Nikolaos, Doss, Srikanth, Sunkara, Monica, Halder, Kishaloy, Mager, Manuel, Benajiba, Yassine
Aligning Large Language Models (LLM) to address subjectivity and nuanced preference levels requires adequate flexibility and control, which can be a resource-intensive and time-consuming procedure. Existing training-time alignment methods require ful
Externí odkaz:
http://arxiv.org/abs/2410.19206
Autor:
Liu, Siyi, Ning, Qiang, Halder, Kishaloy, Xiao, Wei, Qi, Zheng, Htut, Phu Mon, Zhang, Yi, John, Neha Anna, Min, Bonan, Benajiba, Yassine, Roth, Dan
Open domain question answering systems frequently rely on information retrieved from large collections of text (such as the Web) to answer questions. However, such collections of text often contain conflicting information, and indiscriminately depend
Externí odkaz:
http://arxiv.org/abs/2410.12311
Autor:
Liu, Qin, Shang, Chao, Liu, Ling, Pappas, Nikolaos, Ma, Jie, John, Neha Anna, Doss, Srikanth, Marquez, Lluis, Ballesteros, Miguel, Benajiba, Yassine
The safety alignment ability of Vision-Language Models (VLMs) is prone to be degraded by the integration of the vision module compared to its LLM backbone. We investigate this phenomenon, dubbed as ''safety alignment degradation'' in this paper, and
Externí odkaz:
http://arxiv.org/abs/2410.09047
As large language models (LLMs) become increasingly versatile, numerous large scale benchmarks have been developed to thoroughly assess their capabilities. These benchmarks typically consist of diverse datasets and prompts to evaluate different aspec
Externí odkaz:
http://arxiv.org/abs/2410.05952
Autor:
Vacareanu, Robert, Pratik, Anurag, Spiliopoulou, Evangelia, Qi, Zheng, Paolini, Giovanni, John, Neha Anna, Ma, Jie, Benajiba, Yassine, Ballesteros, Miguel
Many of the recent capabilities demonstrated by Large Language Models (LLMs) arise primarily from their ability to exploit contextual information. In this paper, we explore ways to improve reasoning capabilities of LLMs through (1) exploration of dif
Externí odkaz:
http://arxiv.org/abs/2405.00204
Autor:
Wang, Fei, Shang, Chao, Jain, Sarthak, Wang, Shuai, Ning, Qiang, Min, Bonan, Castelli, Vittorio, Benajiba, Yassine, Roth, Dan
User alignment is crucial for adapting general-purpose language models (LMs) to downstream tasks, but human annotations are often not available for all types of instructions, especially those with customized constraints. We observe that user instruct
Externí odkaz:
http://arxiv.org/abs/2403.06326
Autor:
Hwang, Alyssa, Dixit, Kalpit, Ballesteros, Miguel, Benajiba, Yassine, Castelli, Vittorio, Dreyer, Markus, Bansal, Mohit, McKeown, Kathleen
We present NewsQs (news-cues), a dataset that provides question-answer pairs for multiple news documents. To create NewsQs, we augment a traditional multi-document summarization dataset with questions automatically generated by a T5-Large model fine-
Externí odkaz:
http://arxiv.org/abs/2402.18479
Autor:
Chang, Tyler A., Halder, Kishaloy, John, Neha Anna, Vyas, Yogarshi, Benajiba, Yassine, Ballesteros, Miguel, Roth, Dan
NLP models often degrade in performance when real world data distributions differ markedly from training data. However, existing dataset drift metrics in NLP have generally not considered specific dimensions of linguistic drift that affect model perf
Externí odkaz:
http://arxiv.org/abs/2305.17127
Autor:
Lesci, Pietro, Fujinuma, Yoshinari, Hardalov, Momchil, Shang, Chao, Benajiba, Yassine, Marquez, Lluis
Publikováno v:
Findings of the Association for Computational Linguistics: ACL 2023
Sequence-to-sequence state-of-the-art systems for dialogue state tracking (DST) use the full dialogue history as input, represent the current state as a list with all the slots, and generate the entire state from scratch at each dialogue turn. This a
Externí odkaz:
http://arxiv.org/abs/2305.17020