Maitrey Mehta

PhD student at Utah NLP| Incoming Applied Science Intern at Amazon | #NLProc | Low-resource NLP |

[Kenneth Marino @Kenneth_Marino]

Super excited that the Computer Use survey I've been working on w/ @anmarasovic for a while now is ready! Originally we were planning on a more traditional survey paper but as more surveys came out we decided on an interactive website survey.

[Vivek Gupta ✈️ ACL 2026 @keviv9]

#EMNLP2025 results are out! 🎉
We got 5 papers from CoRAL @SCAI_ASU (+collaborators):

➡️ 2 in Main
➡️ 3 in Findings

Huge congratulations to my amazing students (many first-timers 💪) and collaborators! 🥳🍾🎈

#NLP #NLProc #LLM #StructuredData
@ASUEngineering, @ASU (1/n)

[rishanth rajendhran @rishanthrajendh]

Accepted to Findings of #EMNLP2025!

[Maitrey Mehta @my_tray]

@lasha_nlp @mpi_sws_ Congrats Abhilasha!!

[MClem @mclemcrew]

Absolutely stoked to announce that our paper “MixAssist: An Audio-Language Dataset for Co-Creative AI Assistance in Music Mixing” was accepted at COLM’25!

This research aims to foster development of AI that empowers artists by augmenting their skills, not automating their flow.

[Ana Marasović @anmarasovic]

My first audio AI paper, thanks to @mclemcrew who introduced me to a whole new world of music producing! Getting to work with students who bring their unique passion into PhD work is one of the best perks of a professor's job.

Check more in the thread 👇🏻 Soon at #COLM2025

[Fateme Hashemi Chaleshtori @fatemehc__]

1/ 🚨NEW PAPER: "BriefMe: A Legal NLP Benchmark for Assisting with Legal Briefs", accepted to ACL Findings 2025!
We introduce the first benchmark specifically designed to help LLMs assist lawyers in writing legal briefs 🧑‍⚖️

[Alex Gill @alex_gill_nlp]

𝐖𝐡𝐚𝐭 𝐇𝐚𝐬 𝐁𝐞𝐞𝐧 𝐋𝐨𝐬𝐭 𝐖𝐢𝐭𝐡 𝐒𝐲𝐧𝐭𝐡𝐞𝐭𝐢𝐜 𝐄𝐯𝐚𝐥𝐮𝐚𝐭𝐢𝐨𝐧?

I'm happy to announce that the preprint release of my first project is online! Developed with the amazing support of @lasha_nlp and @anmarasovic

(Full link below 👇)

[Ana Marasović @anmarasovic]

I'll talk about measuring faithfulness of verbalized reasoning today at Repl4NLP at *9:45* #NAACL2025

[Vivek Gupta ✈️ ACL 2026 @keviv9]

Join us for the poster, at #emnlp2024

Do LLMs Really Understand Charts?
🔍 TL;DR: VLMs seem accurate on charts, but consistency & robustness? Not so much.
With @AparnaGarimell2, Pritika, @DanRothNLP, and others
🗓️ Session 12: Multimodality & Language Grounding 5
📍 Nov 14 (Thu) 14:00–15:30 (Aparna, Pritika, & I’ll present)

[Fateme Hashemi Chaleshtori @fatemehc__]

So excited to share our work at #EMNLP2024 🥳🌴

Join the poster session to chat about our paper on application-grounded evaluation of explanations and more!💭

🕰 Wed. 10:30-12
🔗 Preprint: https://t.co/vNmdYuxMx0

[Zhichao Xu Brutus @zhichaoxu_ir]

paper link https://t.co/I8nBvEYWzC

I'll be at poster session 09 (Wed 16:00 - 17:30), look forward to talking about dialogues, LLM-as-judge and other fun topics

[Ana Marasović @anmarasovic]

I will be at #EMNLP2024! My student 𝙁𝙖𝙩𝙚𝙢𝙚 𝙃𝙖𝙨𝙝𝙚𝙢𝙞 𝘾𝙝𝙖𝙡𝙚𝙨𝙝𝙩𝙤𝙧𝙞 @fatemehc__ will present "On Evaluating Explanation Utility for Human-AI Decision Making in NLP" in the poster session on 𝗪𝗲𝗱𝗻𝗲𝘀𝗱𝗮𝘆, 𝟭𝟬:𝟯𝟬𝗮𝗺: https://t.co/5pVRm2TjRr 1/

[Maitrey Mehta @my_tray]

@valentina__py Congrats!! 🎉

[Maitrey Mehta @my_tray]

@taguchi_c @davidweichiang Congrats Chihiro!!!

[Maitrey Mehta @my_tray]

@aryaman2020 @ChrisGPotts @jurafsky Congrats!

[Zhichao Xu Brutus @zhichaoxu_ir]

📜Research question: can we define counterfactual explanations in the context of web search❓

In our recent @acm_ictir paper, we conceptualize a new explanation paradigm - providing counterfactual explanation for less-proficient searchers.

Details👇🧵

[Zhichao Xu Brutus @zhichaoxu_ir]

Are compressed LLMs less toxic and biased against different demographic groups❓In this new📜, we study 4 pruning methods and 3 quantization methods and evaluate on 7 bias/toxicity benchmarks. https://t.co/hT97eNuoiz

(Un)surprising answer is: they are not less toxic/biased

[Maitrey Mehta @my_tray]

@valentina__py and I will be presenting this work at 11 AM CST on Monday. Please do stop by our poster!! (Location: Don Diego)

[Maitrey Mehta @my_tray]

New preprint 🚨

"Do LLM predictors provide structurally consistent outputs in the zero- and few-shot regime?"

Our new work "Promptly Predicting Structures: The Return of Inference" shows that they do not, and we show how to fix it.

(1/n) 🧵

[Maitrey Mehta @my_tray]

You like the ease of prompting, but you also like the consistency of structured inference? Don't worry, you can have both!

Joint work with @valentina__py and @viveksrikumar.

Paper: https://t.co/CTmxyRf2xO
Code: https://t.co/CuAOMl4nT6

(n/n) 🧵

[Joe Davison @joeddav]

🤗➕🏔️

@UtahNLP is now organized on the @huggingface hub

I recommend you 👀 it

https://t.co/WbOymbRuQT