Debarati Das
@geekylildeb
Senior Applied Scientist @Microsoft PhD Candidate @UMNComputerSci Working on Socially Aware #NLP #CSS (she/her)
You might like
Thrilled that our paper analyzing Microsoft Copilot for consumer and WildChat-1M logs received a Senior Area Chair highlight award at @emnlpmeeting! I'm excited for work on "LLMs in the wild" and grateful for all the energy and interest we received at the #EMNLP2025!
🤔 We know what people are using LLMs for, but do we know how they collaborate with an LLM? 🔍 In a recent paper we answered this by analyzing multi-turn sessions in 21 million Microsoft Copilot for consumers and WildChat interaction logs: arxiv.org/abs/2505.16023
Debarati Das (@geekylildeb), Chi Le (@elihc_2706), and Karin de Langis will present LawFlow in 10 minutes! Interested in modeling human workflow from lawyers and experts? Talk to them! #COLM2025 arxiv.org/abs/2504.18942
🎉 Excited to share that our LawFlow paper has been accepted to COLM 2025. See you all in Montreal! #COLM2025 #NLP #HumanCenteredAI #FutureofWork
Our team will be presenting LawFlow at #COLM2025 🇨🇦! If you're excited about modeling human reasoning, workflow diversity, and human–AI collaboration in complex tasks, come chat with us! 📍Poster - Tue 11:00 am - 1:00 pm @MinnesotaNlp
🎉 Excited to share that our LawFlow paper has been accepted to COLM 2025. See you all in Montreal! #COLM2025 #NLP #HumanCenteredAI #FutureofWork
Thrilled that this work was accepted to #EMNLP2025! It presents the first large scale analysis of users collaborating with LLMs in-the-wild! Also thrilled that this kind of NLP+HCI work received generous support (and feedback) from reviewers and landed in the top 15% of papers!
🤔 We know what people are using LLMs for, but do we know how they collaborate with an LLM? 🔍 In a recent paper we answered this by analyzing multi-turn sessions in 21 million Microsoft Copilot for consumers and WildChat interaction logs: arxiv.org/abs/2505.16023
🎉 Excited to share that our LawFlow paper has been accepted to COLM 2025. See you all in Montreal! #COLM2025 #NLP #HumanCenteredAI #FutureofWork
🚨 New paper! 🚨 Legal work isn’t a checklist. It’s full of ambiguity, changing goals, and evolving info. So why is legal AI mostly built for static tasks like clause selection or doc search? 💡 We argue: model the process, not just the product. ArXiv : arxiv.org/abs/2504.18942
Our new paper 🚨highlights how users steer AI assistants in real-world writing tasks. This has implications for real-time personalization, realistic session evaluation, and learner-centered alignment!
🤔 We know what people are using LLMs for, but do we know how they collaborate with an LLM? 🔍 In a recent paper we answered this by analyzing multi-turn sessions in 21 million Microsoft Copilot for consumers and WildChat interaction logs: arxiv.org/abs/2505.16023
Thrilled to team up with @UofMNLawSchool again. @Dschwarcz reshaped our #lawflow paper into a sharp, insightful position paper! Building truly cognitively aligned expert-support systems means to consider multidimensional human factors, like seniority, trust, role…
🧵 New Essay: Thinking Like a Lawyer in the Age of Generative AI. papers.ssrn.com/sol3/papers.cf… Despite AI’s power to transform legal work, legal markets have barely budged. Why? Our answer: Lawyers—especially junior ones—struggle to deeply understand AI outputs in complex legal tasks.
7/ Check out the draft, co-authored with @dongyeopkang, @geekylildeb & Brett McDonnell, & feel free to offer suggestions and feedback! papers.ssrn.com/sol3/papers.cf…
This paper introduces LawFlow, a dataset capturing complete legal workflows from law students, enabling comparison with AI and suggesting improved AI assistance. 📌 LawFlow captures 'chain-of-decisions', revealing human legal reasoning's dynamic, iterative, multi-path nature.…
@geekylildeb showed us how to human study workflows and had fascinating insights about how humans work vs. agents. While agents stringently advance the steps of their workflow, humans are more flexible and iterative. Hear this, all ye agent designers!! arxiv.org/html/2504.1894…
Honored to be a speaker today at #WiADS2024 ! It was a great experience discussing the impact of artifacts present in LLM-generated data with this talented crowd! 🙂 #womeninai #nlproc #artifacts #LLMs
United States Trends
- 1. Wemby 32.3K posts
- 2. Steph 68.4K posts
- 3. Spurs 30.9K posts
- 4. Draymond 13.4K posts
- 5. Clemson 11.2K posts
- 6. Louisville 10.9K posts
- 7. #SmackDown 50.9K posts
- 8. Zack Ryder 15.8K posts
- 9. Aaron Fox 2,214 posts
- 10. #DubNation 2,007 posts
- 11. Harden 14K posts
- 12. Brohm 1,655 posts
- 13. Marjorie Taylor Greene 45.2K posts
- 14. Massie 54.4K posts
- 15. Dabo 1,972 posts
- 16. Landry Shamet 5,871 posts
- 17. UCLA 8,568 posts
- 18. Mitch Johnson N/A
- 19. Bill Clinton 184K posts
- 20. Matt Cardona 2,868 posts
You might like
-
Qiao (Georgie) Jin
@QiaoJin5 -
Girl Geek X Community
@GirlGeekX -
Estelle Smith 🌟
@EstelleSmithPhD -
Tigo El Salvador
@TigoElSalvador -
Alex Dimakis
@AlexGDimakis -
neural oscillator of uncertain significance
@mycoliza -
Shubhendu Trivedi
@_onionesque -
Dongyeop Kang (DK)
@dongyeopkang -
Jamie Thomas
@tjamieibm -
Vincent Menezes
@vincentmenezes
Something went wrong.
Something went wrong.