Výsledky vyhledávání - "Serrano, Sofia"

Report

Stubborn Lexical Bias in Data and Models

Autor: Serrano, Sofia, Dodge, Jesse, Smith, Noah A.

In NLP, recent work has seen increased focus on spurious correlations between various features and labels in training data, and how these influence model behavior. However, the presence and effect of such correlations are typically examined feature b

Externí odkaz: http://arxiv.org/abs/2306.02190

Zobrazit plný text záznamu

Report

All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text

Autor: Clark, Elizabeth, August, Tal, Serrano, Sofia, Haduong, Nikita, Gururangan, Suchin, Smith, Noah A.

Human evaluations are typically considered the gold standard in natural language generation, but as models' fluency improves, how well can evaluators detect and judge machine-generated text? We run a study assessing non-experts' ability to distinguis

Externí odkaz: http://arxiv.org/abs/2107.00061

Zobrazit plný text záznamu

Report

Is Attention Interpretable?

Autor: Serrano, Sofia, Smith, Noah A.

Attention mechanisms have recently boosted performance on a range of NLP tasks. Because attention layers explicitly weight input components' representations, it is also often assumed that attention can be used to identify information that models foun

Externí odkaz: http://arxiv.org/abs/1906.03731

Zobrazit plný text záznamu

Akademický článek

Study of an Iron Age Gilded Silver Earring by XRF, SEM-EDS and Multifocus OM.

Autor: Serrano, Sofia¹ (AUTHOR) spe.serrano@campus.fct.unl.pt, Rodrigues, Alexandra² (AUTHOR), Silva, Rui J. C.¹ (AUTHOR), Figueiredo, Elin¹ (AUTHOR) spe.serrano@campus.fct.unl.pt

Publikováno v: Heritage (2571-9408). May2023, Vol. 6 Issue 5, p4187-4201. 15p.