Tanu Mitra
@tanmit
Followers
2K
Following
1K
Media
73
Statuses
770
Assistant Professor @UW_iSchool. Computer Science, PhD @GeorgiaTech Social computing researcher studying online misinformation. Previously @VT_CS, @gtcomputing.
Joined September 2009
Figuring it out what’s it like attending the largest ACL conference. Also returning to attend my first ACL as a faculty. The last one I attended was almost half a decade ago as a PhD student #ACL2025NLP
0
0
9
"A stylish woman in a beige coat walking confidently past a moving train at a modern train station.". Create images and videos in seconds with Grok Imagine.
802
718
4K
Woohoo! Congrats Hayoung! Such a well deserved award. It was a treat to work with you all these years and I can’t wait to see all the awesome things you are going to do next in your PhD journey!.
Life update: I graduated with my M.S. @uwcse and received the outstanding master’s thesis award, advised by @tanmit! . I am truly grateful for the amazing mentorship from Prof. Tanu Mitra and many others who have supported and taught me 💜💛
1
0
7
RT @huashen218: 🚀 BiAlign 🚀 Thrilled to announce that 72 papers have been accepted to our Bidirectional Human-AI Alignment workshop at #ICL….
0
11
0
Starting right now in our RAISE event @ecekamar keynote on AI agents. So much to learn in this fast moving field! @uw_ischool @uwcse
0
0
3
Happening now in Zillow Commons, UW campus - our Winter RAISE exposition event. If you are nearby, do attend.
0
0
0
Interested in Human-AI Alignment research and want to workshop the complex and evolving challenges in this space? Consider submitting your work at our interdisciplinary #ICLR2025 workshop @iclr_conf & #CHI2025 SIG. .Deadline Feb 15th.
bialign-workshop.github.io
The official website for the ICLR BiAlign: Workshop on Bidirectional Human-AI Alignment
1
2
20
RT @snaglee2401: Recommender systems can learn from how r\ifyoulikeblank users solicit recommendations from each other. Our @ACM_CSCW paper….
dl.acm.org
Despite the existence of search-based recommender systems like Google, Netflix, and Spotify, online users sometimes may turn to crowdsourced recommendations in places like the r/ifyoulikeblank...
0
4
0
RT @preetams7: 2️⃣ Unveiling Covert Harms in LLM Conversations.📍 Brickell | Nov 14 | 10:30-12 PM.🔗 (W/ @hayounggju….
aclanthology.org
Preetam Prabhu Srikar Dammu, Hayoung Jung, Anjali Singh, Monojit Choudhury, Tanu Mitra. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.
0
3
0
RT @amyjko: The @UW_iSchool is hiring this year! Tenure track open area, tenure track LIS, teaching track data science, teaching track LIS.….
ischool.uw.edu
The University of Washington is committed to being one of the top information schools in the world. As part of this effort, we seek outstanding individuals...
0
39
0
RT @joewalther: @TheLabBKC @BKCHarvard “Information Sharing and Content Framing across Multiple Platforms and Functional Roles That Exempli….
0
3
0
I am at #FaCCT2024 for the next few days and can chat in-person too! Bummed to be missing #ICWSM2024 which is happening at the same time, but if folks there are interested, please reach out!!.
0
0
2
If you're a recently graduated or about-to-graduate PhD looking for a postdoc in the space of auditing AI systems, epistemic risks of generative AI, and thinking about human-centric ways of addressing these risks, please apply and reach out:
1
36
80
Robert presenting this right now! @wolferobert3 #FAccT2024
If you are #FAccT2024, don't miss.@wolferobert3.talk today at the 10:10am session on "When LLMs are being deployed". The work highlights the ways fact-checking orgs are using generative AI and the challenges that prevent further adoption. @uw_ischool
0
0
3
This was a fun talk to hear #facct2024 on Evaluation of LLMs' Humour Alignment with Comedians, which essentially ended with "We need more HCI methods" :-)
1
0
19
If you are #FAccT2024, don't miss.@wolferobert3.talk today at the 10:10am session on "When LLMs are being deployed". The work highlights the ways fact-checking orgs are using generative AI and the challenges that prevent further adoption. @uw_ischool
0
1
7
At #ICWSM2024? Be sure to catch @shruti_ph from presentation of our work done @uw_ischool this Wed at 11am. Nicely timed research as we are right in the midst of the Indian elections! Paper pdf here:
0
2
18
This was a fun collaboration with some amazing students @uw_ischool @uwcse and with @monojitchou from @mbzuai. Checkout the tweet thread from @preetams7 summarizing our findings about subtle harms in #LLMs and how they manifest for caste vs. race. @uwresponsibleai.
LLMs are known to generate harmful views, but what are the various and potentially covert forms of harm and identity threats in LLM-generated conversations?. We explore this with @hayounggjung, @anjali_singh35, @monojitchou, and @tanmit. Preprint:
1
3
19
Pleasantly surprised at how effectively @Trevornoah explains of what AI can and cannot do at .@Microsoft Global Nonprofit Leaders Summit #nonprofitsummit. "AI is not doing the thinking. It's the experts who have already done the thinking. "
1
3
23
+100. When will we stop questioning generalizability of studies done in Global South? When are we going to value working with communities in the Global South?.
As someone who happens to do research in not Global North/Western contexts AND who just read thousands of conference paper reviews as a planner - you would be SHOCKED at the ways that reviewers talk down to authors that they believe are from the Global South. It is so fucked up.
2
1
27