Sharon Levy Profile
Sharon Levy

@sharonlevy21

Followers
1,289
Following
582
Media
10
Statuses
115

Incoming Assistant Professor @RutgersCS Fall 2024 | Current Postdoc @jhuclsp PhD from UCSB @ucsbNLP Research in Responsible NLP

Joined June 2019
Don't wanna be here? Send us removal request.
Explore trending content on Musk Viewer
@sharonlevy21
Sharon Levy
1 year
🎉Excited to announce that I will be starting as an Assistant Professor of Computer Science at Rutgers University in Fall 2024! In the meantime, I will spend the next year at Johns Hopkins @jhuclsp as a postdoc with @mdredze !
40
29
315
@sharonlevy21
Sharon Levy
2 years
I’m on the faculty job market! My research focuses on Responsible AI and investigates undiscovered and underrepresented ethical issues in NLP across fairness, trustworthiness, and safety. If you have relevant positions, let me know! Please share/RT!
9
51
233
@sharonlevy21
Sharon Levy
9 months
✨I am hiring PhD students @RutgersCS for Fall 2024!✨ If you are interested in working with me on Responsible AI at Rutgers, you can read more at: Also, if you are attending EMNLP and would like to chat about this, please reach out!
3
64
210
@sharonlevy21
Sharon Levy
2 years
How should I cool down boiling oil? DON’T throw cold water on it! But what happens when I pose this to a model?
Tweet media one
1
23
149
@sharonlevy21
Sharon Levy
1 year
I will be at #ACL2023NLP next week! I am recruiting Ph.D. students in the next academic cycle, so if you are interested in working with me and would like to connect at the conference, please reach out! Looking forward to meeting everyone🎉
@sharonlevy21
Sharon Levy
1 year
🎉Excited to announce that I will be starting as an Assistant Professor of Computer Science at Rutgers University in Fall 2024! In the meantime, I will spend the next year at Johns Hopkins @jhuclsp as a postdoc with @mdredze !
40
29
315
3
10
116
@sharonlevy21
Sharon Levy
1 year
Excited to share my work from my Amazon internship last summer! We compare biases across English, Spanish, Chinese, Hebrew, and Italian. We analyze biases for race, religion, nationality, and gender attributes on the downstream sentiment analysis task.
Tweet media one
4
23
108
@sharonlevy21
Sharon Levy
11 months
Thank you for the warm welcome! Excited to be part of the team😊
@jhugenderequity
Gender Equity Unit @Johns Hopkins Bloomberg School
11 months
Sharon Levy is a welcome new addition to the team! She is interested in exploring the intersection of public health and AI, with expertise in natural language processing and Responsible AI. She has a PhD in Computer Science from the University of California at Santa Barbara.
Tweet media one
2
3
24
0
0
54
@sharonlevy21
Sharon Levy
9 months
Presenting our work on social biases and the impact of multilingual training at #EMNLP2023 this week! Come check out our poster on Saturday at 11am🎉
@sharonlevy21
Sharon Levy
1 year
Excited to share my work from my Amazon internship last summer! We compare biases across English, Spanish, Chinese, Hebrew, and Italian. We analyze biases for race, religion, nationality, and gender attributes on the downstream sentiment analysis task.
Tweet media one
4
23
108
0
12
54
@sharonlevy21
Sharon Levy
2 years
Excited to share two papers accepted to #EMNLP2022 ! “SafeText: A Benchmark for Exploring Physical Safety in Language Models” w/ Emily Allaway, Melanie Subbiah, @hmslydia , @DrDesmondPatton , Kathleen McKeown, and @WilliamWangNLP accepted to main conference! [1/2]
1
4
48
@sharonlevy21
Sharon Levy
1 year
Thank you William! I've had an amazing experience working with you and @ucsbNLP !
@WilliamWangNLP
William Wang
1 year
🎉 Dr. @sharonlevy21 has successfully defended her Ph.D. on "Responsible AI via Responsible Large Language Models" at UCSB, & bagging the Fiona and Michael Goodchild Graduate Mentoring Award! 🏆 She's #7 PhD from my lab, and #5 to pursue a faculty job. Congrats Dr. Levy! 🚀🎓👏
Tweet media one
0
3
76
1
0
41
@sharonlevy21
Sharon Levy
7 months
It was great visiting the team, thank you for inviting me!
@MegagonLabs
Megagon Labs
7 months
@sharonlevy21 joined us at our office and shared her works in creating responsible language models. ✨Thank you Sharon Levy for an insightful session! #ResponsibleAI #MachineLearning #AI #NLProc #NLP @RutgersU
Tweet media one
0
3
14
0
0
36
@sharonlevy21
Sharon Levy
2 years
Happy to share that our paper “HybriDialogue: An Information-Seeking Dialogue Dataset Grounded on Tabular and Textual Data” was accepted to #ACL2022 Findings! w/ Kai Nakamura, @pascaltuan , @WenhuChen , and @WilliamWangNLP
Tweet media one
0
3
31
@sharonlevy21
Sharon Levy
3 years
Excited to share that our demo paper “Open-Domain Question-Answering for COVID-19 and Other Emergent Domains” was accepted to #EMNLP2021 ! w/ Kevin Mo, @XiongWenhan , and @WilliamWangNLP !
1
2
31
@sharonlevy21
Sharon Levy
2 years
Presenting this at #EMNLP2022 on Friday at 2PM in the Collaboratorium! Looking forward to meeting everyone this week🙂 Let me know if you would like to chat!
@sharonlevy21
Sharon Levy
2 years
How should I cool down boiling oil? DON’T throw cold water on it! But what happens when I pose this to a model?
Tweet media one
1
23
149
2
4
28
@sharonlevy21
Sharon Levy
1 year
Huge thank you to my advisor @WilliamWangNLP and my wonderful colleagues, friends, and family for supporting me through this journey @ucsbNLP @ucsbcs ! I’m recruiting Ph.D. students in the upcoming application cycle. If you’re interested in Responsible NLP, please reach out!✨
2
0
22
@sharonlevy21
Sharon Levy
2 years
“Mitigating Covertly Unsafe Text within Natural Language Systems” w/ @alexmeigz , @aniyipyip , Melanie Subbiah, Emily Allaway, John N. Judge, @DrDesmondPatton , Bruce Bimber, Kathleen McKeown and @WilliamWangNLP accepted to findings! Preprints coming soon! [2/2]
0
3
15
@sharonlevy21
Sharon Levy
2 years
In our #EMNLP 2022 paper, we introduce SafeText, a dataset to benchmark the evaluation of implicitly unsafe text, i.e., text that does not contain explicitly unsafe keywords (e.g. stab) and instead requires additional commonsense knowledge to recognize its physical harm.
Tweet media one
1
2
14
@sharonlevy21
Sharon Levy
3 years
For those of you who missed the #EMNLP2021 demo session, check out our demo paper on open-domain question-answering for COVID-19 and other emergent domains! w/ Kevin Mo, @XiongWenhan , and @WilliamWangNLP
0
2
12
@sharonlevy21
Sharon Levy
3 years
Excited to continue my research on Responsible AI with Amazon!
@WilliamWangNLP
William Wang
3 years
Congratulations to our Ph.D. student Sharon Levy @sharonlevy21 who has been selected to receive an Amazon Fellowship in Responsible AI and #NLProc ! Sharon's #ACL2021nlp paper on memorization of conspiracy theories in large LMs:
0
2
13
0
0
11
@sharonlevy21
Sharon Levy
2 years
Using SafeText, we evaluate several LLM’s for their generation and reasoning skills. We find that models have the capability to generate unsafe text and are not able to reason well between safe and unsafe advice within different scenarios.
1
1
10
@sharonlevy21
Sharon Levy
1 year
Big thank you to my co-authors Neha Anna John, Ling Liu, @yogarshi , Jie Ma, @akkikiki , @migballesteros , Vittorio Castelli, and @DanRothNLP for their help! Check out our paper for more details!
0
0
11
@sharonlevy21
Sharon Levy
2 years
These samples are in the form of a scenario and paired safe and unsafe advice, where the unsafe advice may lead to physical harm.
Tweet media one
1
1
8
@sharonlevy21
Sharon Levy
1 year
To remove ambiguities when translating to grammatically gendered languages, we adapt existing sentiment analysis bias templates in English and explicitly define male/female subjects.
Tweet media one
1
0
8
@sharonlevy21
Sharon Levy
1 year
Our findings show that biases can manifest differently across various languages for individual attributes while also revealing similarities across languages as well such as favoritism of groups dominant in each language’s culture (e.g. majority religions and nationalities).
Tweet media one
1
0
6
@sharonlevy21
Sharon Levy
1 year
We also analyze the impact of multilingual training and how this affects biases! We break this down for pretraining and finetuning data. Our results show bias amplification from multilingual finetuning while multilingual pretraining does not have a consistent effect.
Tweet media one
1
0
6
@sharonlevy21
Sharon Levy
2 years
@gsarti_ @AmazonScience Glad we got to intern together and explore NYC! Safe travels back!!
0
0
5
@sharonlevy21
Sharon Levy
9 months
@fredahshi Thank you! We did not try increasing data size (this was something we were limited to in certain languages) and our hypothesis was that each language/culture has its own societal biases, but it would definitely be interesting to see if biases differ and/or reduce with more data!
0
0
2
@sharonlevy21
Sharon Levy
2 years
@deliprao @AlbalakAlon No but I received the same email😂
1
0
1
@sharonlevy21
Sharon Levy
1 year
@limufar Congrats!!🥳🎉
0
0
1
@sharonlevy21
Sharon Levy
2 years
@AlbalakAlon @deliprao I thought so too😅
1
0
1
@sharonlevy21
Sharon Levy
2 years
@Elena_Kochkina Thank you! This looks very relevant, I'll definitely check this out!
0
0
1