Zobrazeno 1 - 10
of 1 707
pro vyhledávání: '"Lily, H."'
Autor:
Zhang, Lily H., Dadkhahi, Hamid, Finkelstein, Mara, Trabelsi, Firas, Luo, Jiaming, Freitag, Markus
Despite growing interest in incorporating feedback to improve language models, most efforts focus only on sequence-level annotations. In this work, we explore the potential of utilizing fine-grained span-level annotations from offline datasets to imp
Externí odkaz:
http://arxiv.org/abs/2410.16509
Generative models of language exhibit impressive capabilities but still place non-negligible probability mass over undesirable outputs. In this work, we address the task of updating a model to avoid unwanted outputs while minimally changing model beh
Externí odkaz:
http://arxiv.org/abs/2406.13660
Autor:
Chen, Angelica, Malladi, Sadhika, Zhang, Lily H., Chen, Xinyi, Zhang, Qiuyi, Ranganath, Rajesh, Cho, Kyunghyun
Preference learning algorithms (e.g., RLHF and DPO) are frequently used to steer LLMs to produce generations that are more preferred by humans, but our understanding of their inner workings is still limited. In this work, we study the conventional wi
Externí odkaz:
http://arxiv.org/abs/2405.19534
Meta-analysis can be a critical part of the research process, often serving as the primary analysis on which the practitioners, policymakers, and individuals base their decisions. However, current literature synthesis approaches to meta-analysis typi
Externí odkaz:
http://arxiv.org/abs/2308.13514
Publikováno v:
Glioma, Vol 7, Iss 2, Pp 3-9 (2024)
Glioblastoma (GBM) is characterized by a high recurrence rate, significant heterogeneity, and poor prognosis. While there has been a shift in recent years to focus on molecular phenotyping, there are limited data regarding the relationship between th
Externí odkaz:
https://doaj.org/article/c1e09d33f4234997a82422288c6c75a3
Autor:
Zhang, Lily H., Ranganath, Rajesh
Methods which utilize the outputs or feature representations of predictive models have emerged as promising approaches for out-of-distribution (OOD) detection of image inputs. However, these methods struggle to detect OOD inputs that share nuisance v
Externí odkaz:
http://arxiv.org/abs/2302.04132
Permutation invariant neural networks are a promising tool for making predictions from sets. However, we show that existing permutation invariant architectures, Deep Sets and Set Transformer, can suffer from vanishing or exploding gradients when they
Externí odkaz:
http://arxiv.org/abs/2206.11925
Autor:
Brittany Jaso-Yim, Mara Eyllon, Pratha Sah, Mariesa Pennine, George Welch, Keke Schuler, Laura Orth, Heather O'Dea, Elizabeth Rogers, Lily H. Murillo, J. Ben Barnes, Georgia Hoyler, Gabrielle Peloquin, Kevin Jarama, Samuel S. Nordberg, Soo Jeong Youn
Publikováno v:
Internet Interventions, Vol 38, Iss , Pp 100777- (2024)
Background: Less than half of adults with mental health disorders in the United States receive appropriate or timely care. Digital Mental Health Interventions (DMHIs) have the potential to bridge this gap. However, real-world adoption of DMHIs is imp
Externí odkaz:
https://doaj.org/article/5bebeb6a6f4a45e4a81e6159ddbe42d9
Autor:
Jaso-Yim, Brittany, Eyllon, Mara, Sah, Pratha, Pennine, Mariesa, Welch, George, Schuler, Keke, Orth, Laura, O'Dea, Heather, Rogers, Elizabeth, Murillo, Lily H., Barnes, J. Ben, Hoyler, Georgia, Peloquin, Gabrielle, Jarama, Kevin, Nordberg, Samuel S., Youn, Soo Jeong
Publikováno v:
In Internet Interventions December 2024 38
Autor:
Jason B. Colditz, Lily H. Hsiao, Brandon G. Bergman, David W. Best, Eric G. Hulsey, Jaime E. Sidani, Bruce L. Rollman, Kevin L. Kraemer
Publikováno v:
Internet Interventions, Vol 35, Iss , Pp 100708- (2024)
In developing public resources for the Networks Enhancing Addiction Recovery – Forum Activity Roadmap (NEAR-FAR), we completed a systematic observational study of English-language online forums related to recovery from alcohol or other drug addicti
Externí odkaz:
https://doaj.org/article/45a08153ee8f44599785b0f2495b71b5