@akshitajha Profile picture

Akshita Jha

@akshitajha

Ph.D. Candidate @VT_CS (she/her) | Responsible NLP | Prev: @GoogleAI, @iiit_hyderabad, @debian, @linuxfoundation

Similar User
Shruti Rijhwani photo

@shrutirij

Mohit Bansal photo

@mohitban47

Anmol Goel photo

@anmgoel

Avi Sil photo

@aviaviavi__

Subhabrata Mukherjee photo

@subho_mpi

Vinodkumar Prabhakaran photo

@vinodkpg

Kabir photo

@kabirahuja004

Sameer Singh photo

@sameer_

Jiao Sun photo

@sunjiao123sun_

Tuhin Chakrabarty photo

@TuhinChakr

Abhilasha Ravichander photo

@lasha_nlp

Sunipa Dev photo

@sunipa17

Koustuv Saha photo

@kous2v

Fereshte Khani photo

@fereshte_khani

Gaurav Verma photo

@verma22gaurav

Pinned

New #ACL2024 paper alert📢: Introducing ViSAGe: "Visual Stereotypes Around the Globe" - a dataset enabling evaluation of known nationality-based stereotypes in Text-to-Image models. Work w/ @sunipa17, @vinodkpg, @cephaloponderer, Sarah, @shachi_dave, @qadrida, @chandakreddy 1/n

Tweet Image 1

Akshita Jha Reposted

Stellar cast and an extremely relevant+timely topic of discussion ✨ looking forward to attending this💕

Get ready for an insightful panel at the #WiNLP workshop on Nov 15, 2024, from 2:10 PM! Join us for "Global Voices: Celebrating Diversity in Minds, Methods, and Multilingualism!" with superstar panelists @sunipa17 @khyathi_chandu @dirk_hovy @monojitchou @malihealikhani!

Tweet Image 1


Akshita Jha Reposted

And this one too! Another upcoming #acl2024 paper on how various geo-cultural stereotypes are reflected in text-to-image generations, and how persistent they are, for which @akshitajha devised this new eval/analysis called "stereotypical pull". Work led by @akshitajha & @sunipa17

New #ACL2024 paper alert📢: Introducing ViSAGe: "Visual Stereotypes Around the Globe" - a dataset enabling evaluation of known nationality-based stereotypes in Text-to-Image models. Work w/ @sunipa17, @vinodkpg, @cephaloponderer, Sarah, @shachi_dave, @qadrida, @chandakreddy 1/n

Tweet Image 1


Akshita Jha Reposted

SeeGULL is one of highest coverage datasets out there for measuring & mitigating biases for communities across the world. Glad to see its multilingual version accepted at ACL 2024. Highly recommend checking the paper out, and saying hi to @mukBhutani, @sunipa17 if you are there!

🎉 New ACL paper alert 🚨: SeeGULL Multilingual: a Dataset of Geo-Culturally Situated Stereotypes across 23 languages and countries, to detect and mitigate unsafe and unwanted generation by language models. Joint work with @sunipa17, @krob, @vinodkpg, @shachi_dave @aclmeeting

Tweet Image 1


Akshita Jha Reposted

for the past 2 weeks I’ve been reading every abstract from #neurips2023 here are my notes about what I discovered: alexzhang13.github.io/blog/2024/neur… i think everyone can learn something new from this, and I hope this resource is useful!


Akshita Jha Reposted

As a result, no anonymity period will be required for papers submitted for the Feb. 1, 2024 TACL deadline or the Feb. 15, 2024 ARR deadline (aclrollingreview.org/dates). The submissions themselves must still be fully anonymized. (2/4) #NLproc


Akshita Jha Reposted

Come chat with us today at the poster session at #NeurIPS2023 where @vinodkpg will be presenting on behalf of all the authors. In this work we create a resource of stereotypes across diverse identities in India through community engaged methods.

As we build resources for model safety, community engaged methods need to be prioritized. More about our work at arxiv.org/abs/2307.10514, and github.com/google-researc…. Come chat with us at NeurIPS neurips.cc/virtual/2023/p…! @vinodkpg @shachi_dave @dinesh_tewari1 @jayagoyal2



Akshita Jha Reposted

#WiNLP2023 will happen at #EMNLP2023, tomorrow, Dec 6. We are very excited to have put together an excellent event. For those of you who are attending in person, we will be in Leo 1. For virtual sessions we will send out info through Eventbrite. Schedule (Singapore std time)🧵


Akshita Jha Reposted

Applied AI team, Comcast, India, has immediate openings (Fall '23) for internships. Prior publications and strong engineering skills required (recommendation letters expected). Part-time/ remote possible. #vision #nlp #multimodal #genAI #internships linkedin.com/posts/maneesh-…


Akshita Jha Reposted

Excited to reveal our final speaker @trustnlp #ACL2023 Sunipa Dev (@sunipa17) is a Research Scientist at Google at their Responsible AI and Human Centered Technologies organization.

Tweet Image 1

Akshita Jha Reposted

Come chat with @akshitajha tomorrow at ACL 2023 as she presents our work on expanding stereotype resources for model evaluations. You can also catch her at the @GoogleAI demo booth on Tuesday at 10 am!

Check out our #ACL2023 paper presenting SeeGULL: a broad-coverage evaluation dataset of region-based stereotypes to detect (& help mitigate) them being reproduced in NLP models & data. Work w/ @aidaa @chandankreddy @shachi_dave @vinodkpg @sunipa17 arxiv.org/pdf/2305.11840…

Tweet Image 1


I’ll be in Toronto to present our work SeeGULL at #ACL2023. Come check out our poster at 2:00PM ET on July 10th (Frontenac Ballroom and Queen’s Quay). Excited to connect with fellow researchers, and chat about #NLProc, PhD, or just life! #ACL2023NLP

Check out our #ACL2023 paper presenting SeeGULL: a broad-coverage evaluation dataset of region-based stereotypes to detect (& help mitigate) them being reproduced in NLP models & data. Work w/ @aidaa @chandankreddy @shachi_dave @vinodkpg @sunipa17 arxiv.org/pdf/2305.11840…

Tweet Image 1


Akshita Jha Reposted

Excited to share our #ACL2023NLP (#NLProc) paper on "Information Synchronization Across Multilingual Semi-Structured Tables"! 📚🔍 We explore the problem of Information Synchronization across multilingual tables and construct a large-scale dataset InfoSync. #NLPResearch -1/n

Tweet Image 1

Akshita Jha Reposted

Work led by our amazing student researcher @akshitajha (who'll be on the job market soon; keep an eye out :)), and my colleague @sunipa17 (who has many other papers coming out soon on this topic; stay tuned), with awesome collaborators @aidaa @chandankreddy and @shachidave


Loading...

Something went wrong.


Something went wrong.