Zobrazeno 1 - 10
of 276
pro vyhledávání: '"Gormley, Matthew"'
Large language models pretrained on extensive web corpora demonstrate remarkable performance across a wide range of downstream tasks. However, a growing concern is data contamination, where evaluation datasets may be contained in the pretraining corp
Externí odkaz:
http://arxiv.org/abs/2407.08716
Many tasks within NLP can be framed as sequential decision problems, ranging from sequence tagging to text generation. However, for many tasks, the standard training methods, including maximum likelihood (teacher forcing) and scheduled sampling, suff
Externí odkaz:
http://arxiv.org/abs/2406.09393
Autor:
Bertsch, Amanda, Ivgi, Maor, Alon, Uri, Berant, Jonathan, Gormley, Matthew R., Neubig, Graham
As model context lengths continue to increase, the number of demonstrations that can be provided in-context approaches the size of entire training datasets. We study the behavior of in-context learning (ICL) at this extreme scale on multiple datasets
Externí odkaz:
http://arxiv.org/abs/2405.00200
Most pretrained language models rely on subword tokenization, which processes text as a sequence of subword tokens. However, different granularities of text, such as characters, subwords, and words, can contain different kinds of information. Previou
Externí odkaz:
http://arxiv.org/abs/2311.07853
Minimum Bayes Risk (MBR) decoding is a method for choosing the outputs of a machine learning system based not on the output with the highest probability, but the output with the lowest risk (expected error) among multiple candidates. It is a simple b
Externí odkaz:
http://arxiv.org/abs/2310.01387
Autor:
Cheng, Hua, Jafari, Rana, Russell, April, Klopfer, Russell, Lu, Edmond, Striner, Benjamin, Gormley, Matthew R.
We introduce a dataset for evidence/rationale extraction on an extreme multi-label classification task over long medical documents. One such task is Computer-Assisted Coding (CAC) which has improved significantly in recent years, thanks to advances i
Externí odkaz:
http://arxiv.org/abs/2307.03859
Autor:
Mathur, Yash, Rangreji, Sanketh, Kapoor, Raghav, Palavalli, Medha, Bertsch, Amanda, Gormley, Matthew R.
Medical dialogue summarization is challenging due to the unstructured nature of medical conversations, the use of medical terminology in gold summaries, and the need to identify key information across multiple symptom sets. We present a novel system
Externí odkaz:
http://arxiv.org/abs/2306.17384
Since the proposal of transformers, these models have been limited to bounded input lengths, because of their need to attend to every token in the input. In this work, we propose Unlimiformer: a general approach that wraps any existing pretrained enc
Externí odkaz:
http://arxiv.org/abs/2305.01625
Autor:
Glover, John, Fancellu, Federico, Jagannathan, Vasudevan, Gormley, Matthew R., Schaaf, Thomas
Scoring the factuality of a generated summary involves measuring the degree to which a target text contains factual information using the input document as support. Given the similarities in the problem formulation, previous work has shown that Natur
Externí odkaz:
http://arxiv.org/abs/2211.16853
Publikováno v:
Advances in Neural Information Processing Systems, volume 35, 2022, pages 1583-1595
Much recent work has been devoted to the problem of ensuring that a neural network's confidence scores match the true probability of being correct, i.e. the calibration problem. Of note, it was found that training with focal loss leads to better cali
Externí odkaz:
http://arxiv.org/abs/2211.11838