Semantic Scholar Research @ AI2 (@ai2_s2research) 's Twitter Profile
Semantic Scholar Research @ AI2

@ai2_s2research

Research team @allen_ai working on AI, HCI, ML, NLP, accessibility, and comp. social science in support of @SemanticScholar's mission of accelerating science.

ID: 1545541630497009664

linkhttp://research.semanticscholar.org calendar_today08-07-2022 22:55:53

50 Tweet

704 Takipçi

23 Takip Edilen

Yoonjoo Lee (@yoonjoo_le2) 's Twitter Profile Photo

Started my internship in Semantic Scholar at Ai2 this summer (four weeks flew by 😂)! It has been a super exciting time working with Pao Siangliulue, Joseph Chee Chang, Jonathan Bragg, and Matt! Excited to see where our work goes with these amazing people!😀

Joseph Chee Chang (@josephcc) 's Twitter Profile Photo

Looking for paid interview participants :) AI2 Ai2 is creating OLMo - an open & state-of-the-art LLM. If you use LLMs for work or research, we'd love to hear your thoughts on the best ways to provide you access to OLMo. Sign up here to participate: surveymonkey.com/r/Q6VVC8K

Ai2 (@allen_ai) 's Twitter Profile Photo

If you'll be attending #ACL2023 in Toronto, say hi to an AI2 researcher! Here are some of the highlights of papers being presented by our team members: blog.allenai.org/ai2-at-acl-202…

If you'll be attending #ACL2023 in Toronto, say hi to an AI2 researcher! Here are some of the highlights of papers being presented by our team members: blog.allenai.org/ai2-at-acl-202…
Kyle Lo (@kylelostat) 's Twitter Profile Photo

here at #ACL2023! excited to share: 🥐paper by Catherine Chen evaluating layout-infused models on unseen visuospatial document layouts ☕️ongoing work on open source LLMs at Ai2 🍅ongoing work at #hci #nlproc intersection at Semantic Scholar Research @ AI2 🧀bluesky codes dms open

Catherine Chen (@cathychen23) 's Twitter Profile Photo

I’ll be presenting this work at a virtual poster session from 11-12:30 ET today, and also at RepL4NLP and SustaiNLP on Thursday! I’ll be at #ACL2023 all week, come say hi! With @ZejiangS Dan Klein Gabriel Stanovsky Doug Downey Kyle Lo

Maria Antoniak (@maria_antoniak) 's Twitter Profile Photo

Are you attending #ACL2023 and interested in cultural analytics? Joins us on Tuesday for a group conversation! We'll meet during the afternoon coffee hour at 3:45pm on the Lakeview Terrace. Matthew Wilkens (@[email protected]) David Bamman Lucy Li @jmendelsohn2

Tom Hope (@hoper_tom) 's Twitter Profile Photo

Want to find interesting researchers? Discover new authors who work on similar methods or tasks to you? Introducing 🌉Bridger🌉! An NLP-powered author discovery system for sparking new ideas. Ai2 Semantic Scholar. bridger.apps.allenai.org

Daniel Weld (@dsweld) 's Twitter Profile Photo

Great article in Nature today highlighting the promise of AI-powered tools for helping scientific researchers! (of course, they mentioned Semantic Scholar )

Ai2 (@allen_ai) 's Twitter Profile Photo

A primary goal of our OLMo LLM project is to transparently release the artifacts & processes we are using. Today, we're releasing AI2 Dolma, a pertaining dataset of 3T tokens from a diverse mix of web content, academic publications, code, books, and more. blog.allenai.org/dolma-3-trilli…

Luca Soldaini ✈️ ICLR 25 (@soldni) 's Twitter Profile Photo

Announcing Dolma, the dataset for Ai2's LLM, OLMo. It's 3+ trillion tokens (web/papers/code/books/wiki). We hope it will facilitate study of LLMs & their behavior! Released on Hugging Face w ImpACT license huggingface.co/datasets/allen… Overview/datasheet blog.allenai.org/dolma-3-trilli…

Announcing Dolma, the dataset for <a href="/allen_ai/">Ai2</a>'s LLM, OLMo. It's 3+ trillion tokens (web/papers/code/books/wiki). We hope it will facilitate study of LLMs &amp; their behavior!

Released on <a href="/huggingface/">Hugging Face</a> w ImpACT license huggingface.co/datasets/allen…

Overview/datasheet blog.allenai.org/dolma-3-trilli…
Daniel Weld (@dsweld) 's Twitter Profile Photo

Interested in a better way to explore #VLDB2023 papers? Try exp-sum.apps.allenai.org for an LLM-powered way to probe those papers… * Ask questions w/ a single click * Explore answer provenance using the ending “ * Dive deep w/ recursive questions Powered by Semantic Scholar

Ai2 (@allen_ai) 's Twitter Profile Photo

🚨 Awesome new demo from the Semantic Scholar Research @ AI2 team! Explore #VLDB2023 paper abstracts using LLM-powered features with "Expandable Abstracts": exp-sum.apps.allenai.org

🚨 Awesome new demo from the <a href="/ai2_s2research/">Semantic Scholar Research @ AI2</a> team! 
Explore #VLDB2023 paper abstracts using LLM-powered features with "Expandable Abstracts":

exp-sum.apps.allenai.org
Joseph Chee Chang (@josephcc) 's Twitter Profile Photo

When reading an abstract sometime you just want to learn a bit more about the work: What's this "novel design" they proposed? How was the "better accuracy" measured? Try our LLM-powered expandable & explorable summary from Semantic Scholar Research @ AI2 Raymond Fok exp-sum.apps.allenai.org (1/n)

Philipp Mayr (@philipp_mayr) 's Twitter Profile Photo

After a year of silence, the Scholarly Document Processing (SDP) workshop is back. We have been accepted at ACL 2025 #nlproc All information will follow here: 4th Workshop on Scholarly Document Processing (sdproc.org)

After a year of silence, the Scholarly Document Processing (SDP) workshop is back. We have been accepted at <a href="/aclmeeting/">ACL 2025</a> #nlproc

All information will follow here: 4th Workshop on Scholarly Document Processing (sdproc.org)
Maria Antoniak (@maria_antoniak) 's Twitter Profile Photo

New work from me, Aakanksha Naik, Lucy Lu Wang, Irene Chen, and Carla S. Alvarado. We focused on a specific topic — maternal health — to create a set of guiding principles, with grounded advice and questions, for the use of NLP for healthcare applications and research.

Tom Hope (@hoper_tom) 's Twitter Profile Photo

Scimon🧪: Scientific Inspiration Machines Optimized for Novelty. LLMs + inspiration retrieval from papers + 🚀iterative novelty boosting🚀: 1. LLM (e.g. GPT4) generates idea 2. Compares to detected prior work 3. Refines idea novelty 4. Repeat until sufficient novelty reached