Sumit Asthana
@sumit_ast
Followers
290
Following
2K
Media
7
Statuses
221
HCI PhD @ UMich - using technology to make education more accessible, how people's background and cognitive science affects their interaction with techonology.
Joined August 2014
📚How do we make unfamiliar text easy for adult readers? Excited to present my work at #EMNLP2024 Main on "Evaluating LLMs for Targeted Concept Simplification for Domain-Specific Texts" with amazing @hjrashkin @eaclark07 @fantinehuot @mlapata -
2
6
20
RT @TuhinChakr: New paper on human-AI interaction. We hire 18 writers to edit quirks in AI writing & see if #AI can mimic this process to i….
0
41
0
RT @VedantNeedMoEdu: 😓 #AI co-pilots excel in information work, but what about emotional ones? Working with angry clients harms #MentalHea….
0
7
0
RT @singhanj13: New preprint out! . GenAI is changing how we think 🧠 . This work reviews its cognitive impacts. Using Bloom’s Taxonomy & D….
arxiv.org
The rapid adoption of Generative AI (GenAI) is significantly reshaping human cognition, influencing how we engage with information, think, reason, and learn. This paper synthesizes existing...
0
17
0
RT @lal_yash: First CfP for #EMNLP2025 is live now. Submission deadline to May ARR cycle!. Excited to be part of organizing the conference….
0
1
0
RT @huashen218: 🤗Thanks to those asking about our CHI session on Bidirectional 👫Human-AI🤖 Alignment!.🔥Exciting news — We’ll host both the 2….
0
12
0
0
7
0
RT @lucy3_li: New @allen_ai paper accepted to #NAACL2025! We worked with @TeachingLabHQ and @assistments to create a new dataset for evalua….
0
23
0
RT @shanestorks: How well can VLMs detect and explain humans' procedural mistakes, like in cooking or assembly?.🧑🍳🧑🔧. My new pre-print wi….
0
6
0
RT @xuwanghci: 📢I’m recruiting Ph.D students this cycle @UMichCSE , to invent and study mechanisms to support human expertise development,….
0
66
0
RT @agrima_seth: Honored and humbled to awarded the 2024 Google PhD Fellow in #NLProc. Thankful to my advisor @kentarotoyama, my mentors &….
0
4
0
RT @eaclark07: Virtual poster session is at 9am, and the in-person workshop starts at 10am. See you there! #EMNLP2024.
0
3
0
RT @lal_yash: #EMNLP2024 make sure to come to Ashe Auditorium at 5pm today to learn more about 😺 CaT-Bench! Great way to end the conference….
0
6
0
RT @benzevgreen: Can laws be translated into software? In a new study, .@nelescher @jeffbilik @nikola_banovic and I show that the translati….
dl.acm.org
Proponents of legal automation believe that translating the law into code can improve the legal system. However, research and reporting suggest that legal software systems often contain flawed...
0
18
0
RT @cmalaviya11: Excited to share ✨ Contextualized Evaluations ✨!. Benchmarks like Chatbot Arena contain underspecified queries, which can….
0
31
0
Stop by the Generation and Summarization track at 2:00 pm today to learn more about how to make it easy for adult readers to read domain-specific text containing unfamiliar concepts from our talk!.
0
0
0
RT @TuhinChakr: I will be at @emnlpmeeting in Miami presenting our Main conference paper on how Human & AI perform on cognitively demandin….
0
12
0
I would love to chat with people at # EMNLP2024 in Miami about using NLP tools to make educational content accessible, human-AI interaction, understanding how users' backgrounds shape their interactions with AI, and any other fun topics!.
0
0
0
LLMs struggle to balance contextually explaining difficult concepts, preserving the ideas in the definitions, and identifying the right level of explanation for readers. Automated metrics also show low correlations with human-evaluation simplification understanding.
1
0
0
We introduce a dataset of 22k concept definitions from 13 academic domains, paired with an unfamiliar concept in the definition for lay readers, and evaluate four popular LLMs for their ability to simplify concepts in difficult text using human-evaluations and automated measures.
1
0
0
To support this, we introduce the task of "Targeted concept simplification," which can help readers understand unfamiliar, difficult concepts through simplifications or explanations without reducing the overall complexity and nuance of the text.
1
0
0