🎉SAC Awards: 8) COKE: A Cognitive Knowledge Graph for Machine Theory of Mind by Wu et al. 9) MIDGARD: Self-Consistency Using Minimum Description Length for Structured Commonsense Reasoning by Nair and Wang #NLProc #ACL2024NLP

👏A round of applause to PhD student Inderjeet Jayakumar Nair and Prof. Lu Wang on winning an🏆SAC Area Chair's Award - announced today at #ACL2024! Awarded to only 21 publications of the 1915 main conference papers accepted. @ACLmeeting 👉Check out their paper: aclanthology.org/2024.acl-long.…

🎉 Huge congratulations to PhD student Inderjeet Jayakumar Nair & Prof. Lu Wang on winning an SAC Area Chair's Award at #ACL2024! So well-deserved—cheers to your continued success!🌟👏👏 aclanthology.org/2024.acl-long.…

Current LLMs use RLHF to reduce explicit bias in their outputs. But do they also address *implicit* bias? In our recent work, we identify the presence of implicit bias in multi-agent LLM interactions and propose strategies to address it. [1/8]

🌍 How Verifiable Are LM Responses in the Wild? A Three-Way Factuality Benchmark Meet 𝐅𝐚𝐜𝐭𝐁𝐞𝐧𝐜𝐡 – an updatable benchmark for evaluating language models' factuality in real-world scenarios. 🔗 huggingface.co/spaces/launch/… LaunchNLP MichiganAI Computer Science and Engineering at Michigan

Heard of the Alaska-Hawaii merger?🤔Wonder if LLMs know it’s pending government approval before it can happen? They stumble, but we’ve got a fix⚒️! Dive into my #EMNLP2024 work 𝐍𝐚𝐫𝐫𝐚𝐭𝐢𝐯𝐞-𝐨𝐟-𝐓𝐡𝐨𝐮𝐠𝐡𝐭—a special prompting technique to unlock LLMs’ temporal reasoning

Excited to present this work at #EMNLP2024 on Tuesday, 11/12, 14:00-15:30 poster session at Riverfront Hall. I'll be discussing multi-agent LLMs, biases, and mitigation considerations. Looking forward to engaging with you all! 🤗

Large Multimodal Models (LMMs) are unlocking the potential for culture-aware image captions! 🌍 Check out this new work by Longju Bai,Angana Bora, Oana Ignat 👩‍💻🎓📚🇷🇴🌍, and Rada Mihalcea:🔽

PhD student Inderjeet Nair has received a 🏆 Rackham International Student Fellowship for his outstanding academic record. This award will support his continued research on advanced applications of large language models. 👏 Congrats Inderjeet! 👇More: cse.engin.umich.edu/stories/inderj…

This work has been accepted to NAACL 2025 (Main)! 😊 Check out our paper here: arxiv.org/pdf/2411.11758

I had an incredible experience at the Ecosystems, Finance & Health Innovation Sprint 2025 (ecosystemsfinancehealth.org/hackathon-2025) in Nairobi, Kenya. #efhhackathonsprint2025 #aikenya #michiganai #ai #llm #llmagents #rag #climatechange #health #biodiversity

Do LLMs exhibit belief-based biases like humans do? We examine the presence of belief congruence in multi-agent systems and find that LLMs exhibit even stronger belief congruence than humans. Paper: arxiv.org/pdf/2503.02016 Code: github.com/MichiganNLP/Be… 🧵(1/n)

Last week, I had a fantastic time presenting our work on belief congruence in LLMs at the Midwest Speech and Language Days (MSLD) 2025, hosted at Notre Dame.💡 Grateful to the organizers for putting together such a great event! 🙏✨Notre Dame CSE

So honored to have won the Outstanding Paper award at C3NLP today. ❤️ Many thanks to my wonderful collaborators Aparna Garimella and Rada Mihalcea! What an amazing way to end #NAACL2025! 😊

🔍LLMs now give medical diagnoses, legal advice, and even tackle scientific problems. ❓Your LLM sounds smart. But what if it’s just good at faking expertise? 🚀We built ExpertLongBench to find out. 📉And the results? They revealed several concerns.👇 🔗 huggingface.co/spaces/launch/…

Excited to be in Vienna 🇦🇹 next week for #ACL2025 to present our work on simulating a social psychology theory with LLMs and its impact on misinformation and learning. Let’s connect if you're into multi-agent LLMs, misinformation or social NLP in general! 🤗

Presenting this today ACL 2025 🤗 Join me at Poster in Hall X4 Board 3 from 11:00–12:30pm CET to chat about using LLM-based social simulations to explore belief congruence theory and its effects on misinformation dissemination and learning. #ACL2025NLP

I had an incredible experience at #ACL2025NLP in Vienna 🇦🇹! Some highlights 💫: 1. We presented ‘Mind the (Belief) Gap: Group Identity in the World of LLMs’ at the Findings poster session: aclanthology.org/2025.findings-… Huge thanks to my collaborators Marwa Houlla and Rada Mihalcea 🤗