RT @youzhiwen_: (1/4) .Accurate health info matters!. Detecting hallucinated but factual content in plain language summaries?. Introducing….

RT @lifan__yuan: Wanna train PRMs but process labels, annotated manually or automatically, sound too expensive to you😖? .Introduce Implicit….

RT @iSchoolUI: Yue Guo has joined the #iSchoolUI as an assistant professor. She is interested in advancing healthcare delivery by integrati….

Excited to share that our paper on plain language summarization evaluation has been accepted to the #EMNLP2024 main conference! I’ll be in Miami and will have several PhD openings for Fall 2025. Feel free to reach out if you’d like to chat!.

RT @MinyangTian1: SciCode is our new benchmark that challenges LMs to code solutions for scientific problems from advanced papers. The chal….

@UofIllinois Huge thank to my wonderful mentors: Trevor @UW, Lucy @lucyluwang, Gondy @GondyLeroyUA, Sheng @wangshengpkucn, and Tal @tal_august 🥰🥰🥰.

I'm joining the UIUC @UofIllinois this fall as an Assistant Professor in the iSchool, with an affiliation in Computer Science!. My research passion lies in the intersection of NLP and the medical domain. I'm recruiting students for 2025! Check more info:

RT @lifan__yuan: Introducing 🚀Eurus, a suite of state-of-the-art LLM reasoning generalists powered by a new member of Ultra-Series, UltraIn….

RT @tal_august: I will also be at #EMNLP2023 this week. If you’re interested in learning more, I would love to talk!.

(7/n) Our work is an exciting step to integrating an individual scientist's knowledge into jargon detection and to supporting researchers in exploring ideas outside of their domain!.

(6/n) We dive into baselines for predicting term familiarity using a researcher's own papers, including regression models and prompt-based approaches. Our analysis reveals that incorporating individual-level information improves the accuracy of models.

(5/n) Some interesting nuances across domains: generally the humanities were more familiar than the sciences, but annotators wanted examples to explain terms a similar amount across fields!

(4/n) To make progress on this task, we gather a dataset 10k+ term familiarity annotations from CS researchers reading 100 abstracts in 22 domains (from Art to Chemistry).

(3/n) We propose the task of predicting the familiarity of a term for an individual researcher. We ground our task in an initial user study where researchers read papers in less familiar domains and identified their information needs.

(2/n)Scientific jargon helps in-field communication but is tough to grasp across different fields. Traditional methods use general corpora like Wikipedia, but what if we personalize it based on individual scientists' unique backgrounds?.

(1/n) New work! 👏👏. Personalized Jargon Identification for Enhanced Interdisciplinary Communication. My intern project @SemanticScholar @allen_ai with @tal_august, @lucyluwang, Trevor Cohen@UWBIME, @josephcc, @maria_antoniak, Erin Bransom #NLProc

The study will take no more than 15 minutes, and will be conducted asynchronously via a google form. Participants will receive $10 USD upon completion of the study. If you’re interested, please sign up here: Thank you!.

Hello! We are researchers from UW and AI2 looking for participants in a human eval about how AI might improve the way scientists read interdisciplinary paper abstracts. We are recruiting for CS or any related field researchers.

Absolutely phenomenal internship! The team is incredibly welcoming, mentors are a rock, and I've made some awesome new friends for life! 🌟.

RT @marissa_rad: Want to write a blog post for a paper but need a jump-start?.We’re conducting a study to investigate how researchers work….