Zobrazeno 1 - 10
of 19
pro vyhledávání: '"Stolfo, Alessandro"'
The ability to follow instructions is crucial for numerous real-world applications of language models. In pursuit of deeper insights and more powerful capabilities, we derive instruction-specific vector representations from language models and use th
Externí odkaz:
http://arxiv.org/abs/2410.12877
Autor:
Stolfo, Alessandro, Wu, Ben, Gurnee, Wes, Belinkov, Yonatan, Song, Xingyi, Sachan, Mrinmaya, Nanda, Neel
Despite their widespread use, the mechanisms by which large language models (LLMs) represent and regulate uncertainty in next-token predictions remain largely unexplored. This study investigates two critical components believed to influence this unce
Externí odkaz:
http://arxiv.org/abs/2406.16254
Autor:
Stolfo, Alessandro
We present an empirical study of groundedness in long-form question answering (LFQA) by retrieval-augmented large language models (LLMs). In particular, we evaluate whether every generated sentence is grounded in the retrieved documents or the model'
Externí odkaz:
http://arxiv.org/abs/2404.07060
Autor:
Opedal, Andreas, Stolfo, Alessandro, Shirakami, Haruki, Jiao, Ying, Cotterell, Ryan, Schölkopf, Bernhard, Saparov, Abulhair, Sachan, Mrinmaya
There is increasing interest in employing large language models (LLMs) as cognitive models. For such purposes, it is central to understand which properties of human cognition are well-modeled by LLMs, and which are not. In this work, we study the bia
Externí odkaz:
http://arxiv.org/abs/2401.18070
Autor:
Hou, Yifan, Li, Jiaoda, Fei, Yu, Stolfo, Alessandro, Zhou, Wangchunshu, Zeng, Guangtao, Bosselut, Antoine, Sachan, Mrinmaya
Recent work has shown that language models (LMs) have strong multi-step (i.e., procedural) reasoning capabilities. However, it is unclear whether LMs perform these tasks by cheating with answers memorized from pretraining corpus, or, via a multi-step
Externí odkaz:
http://arxiv.org/abs/2310.14491
Mathematical reasoning in large language models (LMs) has garnered significant attention in recent work, but there is a limited understanding of how these models process and store information related to arithmetic tasks within their architecture. In
Externí odkaz:
http://arxiv.org/abs/2305.15054
Step-by-step reasoning approaches like chain of thought (CoT) have proved to be very effective in inducing reasoning capabilities in large language models. However, the success of the CoT approach is fundamentally tied to the model size, and billion
Externí odkaz:
http://arxiv.org/abs/2212.00193
We have recently witnessed a number of impressive results on hard mathematical reasoning problems with language models. At the same time, the robustness of these models has also been called into question; recent works have shown that models can rely
Externí odkaz:
http://arxiv.org/abs/2210.12023
Autor:
Shridhar, Kumar, Monath, Nicholas, Thirukovalluru, Raghuveer, Stolfo, Alessandro, Zaheer, Manzil, McCallum, Andrew, Sachan, Mrinmaya
Ontonotes has served as the most important benchmark for coreference resolution. However, for ease of annotation, several long documents in Ontonotes were split into smaller parts. In this work, we build a corpus of coreference-annotated documents of
Externí odkaz:
http://arxiv.org/abs/2210.03650
Mathematical reasoning in large language models (LLMs) has garnered attention in recent research, but there is limited understanding of how these models process and store information related to arithmetic tasks. In this paper, we present a mechanisti
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4d4975d39711154bca90b6646b869c1f
http://arxiv.org/abs/2305.15054
http://arxiv.org/abs/2305.15054