Tatsunori Hashimoto

Tatsunori Hashimoto

Auto-generated profile, based on publicly available information

Tatsunori Hashimoto

Stanford University, USA

Research Focus & Metrics

Academic Career Length

35Years (1992-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 June 2023
Verified
Conference Paper
Open access
FBFederico Bianchi
Federico Bianchi
PKPratyusha Kalluri
Pratyusha Kalluri
EDEsin Durmus
Esin Durmus
FLFaisal Ladhak
Faisal Ladhak
MCMyra Cheng
Myra Cheng
ACAylin Caliskan
Aylin Caliskan
Machine learning models that convert user-written text descriptions into images are now widely available online and used by millions of users to generate millions of images a day. We investigate the potential for these models to amplify dangerous and complex stereotypes. We find a broad range of ordinary prompts produce stereotypes, including prompts simply mentioning traits, descriptors, occupations, or objects. For example, we find cases of prompting for basic traits or social roles resulting in images reinforcing whiteness as ideal, prompting for occupations resulting in amplification of racial and gender disparities, and prompting for objects resulting in reification of American norms. Stereotypes are present regardless of whether prompts explicitly mention identity and demographic language or avoid such language. Moreover, stereotypes persist despite mitigation strategies; neither user attempts to counter stereotypes by requesting images with specific counter-stereotypes nor institutional attempts to add system “guardrails” have prevented the perpetuation of stereotypes. Our analysis justifies concerns regarding the impacts of today’s models, presenting striking exemplars, and connecting these findings with deep insights into harms drawn from social scientific and humanist disciplines. This work contributes to the effort to shed light on the uniquely complex biases in language-vision models and demonstrates the ways that the mass deployment of text-to-image generation models results in mass dissemination of stereotypes and resulting harms.

14 December 2023 in Foundations and Trends® in Privacy and Security
Verified
Journal Article
Open access
PDF file
CBClark Barrett
Clark Barrett
BBBrad Boyd
Brad Boyd
EBElie Bursztein
Elie Bursztein
NCNicholas Carlini
Nicholas Carlini
BCBrad Chen
Brad Chen
DYDiyi Yang
Diyi Yang
Every major technical invention resurfaces the dual-use dilemma—the new technology has the potential to be used for good as well as for harm. Generative AI (GenAI) techniques, such as large language models (LLMs) and diffusion models, have shown remarkable capabilities (e.g., in-context learning, code-completion, and text-to-image generation and editing). However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks. This monograph reports the findings of a workshop held at Google (co-organized by Stanford University and the University of Wisconsin-Madison) on the dual-use dilemma posed by GenAI. This work is not meant to be comprehensive, but is rather an attempt to synthesize some of the interesting findings from the workshop. We discuss short-term and long-term goals for the community on this topic. We hope this work provides both a launching point for a discussion on this important topic as well as interesting problems that the research community can work to address.

1 June 1992 in Japanese Journal of Applied Physics
Verified
Journal Article
Closed access
JYJiro Yoshida
Jiro Yoshida
THTatsunori Hashimoto
Tatsunori Hashimoto
SIShinji Inoue
Shinji Inoue
YMYoshihisa Mizutani
Yoshihisa Mizutani
MSMasayuki Sagoi
Masayuki Sagoi
KMKoichi Mizushima
Koichi Mizushima
A comprehensive study was carried out on the Josephson effect in YBCO/Au-Ag/Pb SNS junctions prepared on YBCO thin films with various crystal orientations. Most of the fabricated junctions showed a dc Josephson current with a clear Fraunhofer pattern dependence on the applied magnetic field, which ensured the uniformity of the fabricated junction structure. Experimental results were quantitatively compared with the conventional junction theory based on the Usadel equation. The junction characteristics were confirmed to be reasonably explained by a junction model in which a thin insulating or highly resistive layer was assumed to exist at the YBCO-Au interface.

1 January 2023
Verified
Conference Paper
Open access
PDF file
FLFaisal Ladhak
Faisal Ladhak
EDEsin Durmus
Esin Durmus
THTatsunori B. Hashimoto
Tatsunori B. Hashimoto
Faisal Ladhak, Esin Durmus, Tatsunori Hashimoto. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

1 January 2021
Verified
Conference Paper
Open access
PDF file
SMShikhar Murty
Shikhar Murty
THTatsunori B. Hashimoto
Tatsunori B. Hashimoto
CMChristopher D. Manning
Christopher D. Manning
Shikhar Murty, Tatsunori B. Hashimoto, Christopher Manning. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

5 March 2019 in arXiv
Verified
Preprint
Open access
PDF file
EPEmma Pierson
Emma Pierson
PKPang Wei Koh
Pang Wei Koh
THTatsunori Hashimoto
Tatsunori Hashimoto
DKDaphne Koller
Daphne Koller
JLJure Leskovec
Jure Leskovec
PLPercy Liang
Percy Liang
Modeling how individuals evolve over time is a fundamental problem in the natural and social sciences. However, existing datasets are often cross-sectional with each individual observed only once, making it impossible to apply traditional time-series methods. Motivated by the study of human aging, we present an interpretable latent-variable model that learns temporal dynamics from cross-sectional data. Our model represents each individual's features over time as a nonlinear function of a low-dimensional, linearly-evolving latent state. We prove that when this nonlinear function is constrained to be order-isomorphic, the model family is identifiable solely from cross-sectional data provided the distribution of time-independent variation is known. On the UK Biobank human health dataset, our model reconstructs the observed data while learning interpretable rates of aging associated with diseases, mortality, and aging risk factors.