First GLAMOR Lab lunch today since we moved from USC to Georgia Tech! We're so excited to be in Atlanta and look forward to working with our new colleagues in the School of Interactive Computing 😁

Over half of the papers in my ARR batch as AC this cycle contained easily identified, hallucinated citations with incorrect author lists and malformed text. I am so disappointed by the state of the field right now, and I encourage ARR to move forward with desk rejections. #nlproc

This fall semester, I will be moving my GLAMOR lab (glamor.rocks) to the School of Interactive Computing (IC) at the Georgia Institute of Technology!

Okay y’all, gather round for a chat. It’s been a roller coaster, and I thought I’d share what we’ve learned. 🧵 (1/16) bsky.app/profile/luck...

Dr. Lucky Tran@luckytran.com · last yr.

BREAKING: Scientists are staging a “science fair” in the lobby of a Congressional building to tell elected officials about the critical knowledge the US will lose because their research grants have been canceled.

This month, @jessezhang.bsky.social completed his PhD defense and signed to start a postdoc with @abhishekunique7.bsky.social at UW! Keep an eye on his journey :) www.jessezhang.net I'm sad to lose one of my sinistral students but glad to produce another Dr. Jesse 😛

Come by @naaclmeeting.bsky.social Hall 3 posters at 11am to see @leekezar.bsky.social's American Sign Language Knowledge Graph! Lee's work highlights quantitative evidence of iconic relationships between form and meaning in ASL. The ASLKG is part of an exciting direction for the GLAMOR lab :)

Come by @naaclmeeting.bsky.social Poster 6 in Hall 3 from 4-530pm today to see @billzhu.bsky.social's and Ishika Singh's work with me and @robinjia.bsky.social on PSALM: autonomously inducing symbolic pre- and post-conditions of actions with LLMs, symbolic planning, and text environment interaction!

At @naaclmeeting.bsky.social this week! I’ll be presenting our work on LLM domain induction with @thomason.bsky.social on Thu (5/1) at 4pm in Hall 3, Section I. Would love to connect and chat about LLM planning, reasoning, AI4Science, multimodal stuff, or anything else. Feel free to DM!

Human-AI decision making is probably something I think is bad but inevitable, so the least we can do is explore ways to reduce inappropriate human reliance on AI system output. In @tejassrinivasan.bsky.social latest work, he does just that, successfully mitigating /both/ over- and under-reliance.

Tejas Srinivasan@tejassrinivasan.bsky.social · 2y ago

People are increasingly relying on AI assistance, but *how* they use AI advice is influenced by their trust in the AI, which the AI is typically blind to. What if they weren’t? We show that adapting AI assistants' behavior to user trust mitigates under- and over-reliance! arxiv.org/abs/2502.13321

‼️ Ever wish LLMs would just... slow down for a second? In our latest work, "Better Slow than Sorry: Introducing Positive Friction for Reliable Dialogue Systems", we delve into how strategic delays can enhance dialogue systems. Paper Website: merterm.github.io/positive-fri...

The CoRL party continues tomorrow! My student Ishika Singh is helping to organize LangRob (sites.google.com/view/langrob...) where I'll give the last invited talk at 1600, and she is presenting our ongoing work on symbolic planning and LLMs (arxiv.org/abs/2406.02791) at LEAP. Come chat with us!

Trying out a Role-Playing Paper-Reading Seminar in the style of Colin Raffel's blog colinraffel.com/blog/role-pl... post in my History of Language and Computing graduate course this semester. Eager to see how it plays out, but I wanted to show off the class materials that just arrived :)

-Tomorrow- at #NeurIPS2023 you can catch me in Hall B2 at the Robot Learning Workshop bright and early at 8:45am-9:10am talking about how "LPTMs Can Help Robots Without Ignoring Robotics", and where I feel the community (and my own lab!) have mostly missed the mark here so far.

happy spotify wrapped season to the three other people paying for YouTube Music

It's National First-Generation College Student Day! If you navigated part time jobs, college applications and fees, college stress, maybe even grad school all while having to make up stories that explain what you're doing to family, I see you. I'm proud of you, and I'm happy to talk. #FirstGenDocs

Proud of this new work with @billzhu.bsky.social and @robinjia.bsky.social that wades into the blurred space between semantic parsing and open vocabulary generation, to appear at #EMNLP #EMNLP2023 !

Wang Bill Zhu@billzhu.bsky.social · 3y ago

✨ Excited to share our Chain-of-Questions paper #EMNLP2023: we develop a framework that trains *one T5 model* to robustly answer multistep questions by generating and answering sub-questions. Outperforms ChatGPT on DROP, HotpotQA and their contrast/adversarial sets.

I'm at CoLLAs in Montreal this week where my students Tejas (PhD) and Flora (ugrad) are presenting their work on Adapter distillation---transfer learning without the parameter cost of AdapterFusion at tomorrow's poster session board P7. Looking forward to meeting folks in this emerging community 🐨

petition to rename "Postdoc" to "Assistant to the Professor"