✨I’m on the academic job market!✨ I'm a CS Postdoc at Stanford in the Stanford NLP Group. I develop AI and causal inference to tackle societal issues like online civility, racial justice, sustainability, and more. Check out the thread for details! Website: kristinagligoric.com

🤔 What is the most effective metric for summarization? Is it 1⃣ BLEU, 2⃣ ROUGE, or perhaps 3⃣ METEOR? ❔How can we develop a metric that aligns closely with human preferences? ✨We present MetaMetrics, a calibrated meta-metric specifically designed to evaluate generation

Thanks for sharing our latest work on token-level reward models (TLDR) for multimodal models. Paper is out here: arxiv.org/abs/2410.04734

🛎 Thrilled to share that our LLM Persona survey paper is accepted to #EMNLP 2024 Findings and CustomNLP4U 2024 workshop! Hooray! 🎊 See you in Miami~✈️ 🧵 (1/6)

Just published in nature! Research by Google DeepMind including Nobel Laureate Demis Hassabis introduces a novel technique for reliably identifying LLM-generated content. At its core, this approach subtly alters the next-token sampling process to introduce minor adjustments

I was a 2-nd year PhD student when I decided to write the survey paper on bias and fairness in ML. Back then I did it mostly as a literature review for myself and others to learn about the newly emerged field. I never ever imagined that the paper would get the love it got!

Heading to Miami for #EMNLP2024! I'm presenting this paper in the poster session on Thursday afternoon. If you are there and would like to chat, just let me know! 😊

We show that LLMs’ political ideology can be easily manipulated by just a few synthetic samples! 😈 Join Kai Chen on Nov 12 (Tue) from 16:00 to 17:30 in poster session 4 for this exciting paper. #EMNLP2024

It feels emotional to hear that #EMNLP is going back to China after 10 years🥹🥹🥹 thanks EMNLP 2025 ❤️❤️❤️

Our team (Google Cloud AI Research: research.google/teams/cloud-ai/) is seeking PhD student researchers/interns to work on LLM-related topics (agent, reasoning, RAG, data synthesis, etc.). If interested, please fill in this form: forms.gle/Cia2WGY94zTkpP…. Thank you and plz help RT!

🙌 WTG to ALL🏆 the USC Viterbi School students at #EMNLP2024 including ISIers @rebeccadorn_ Brihi Joshi Alexander Spangher (going to NAACL) Tenghao Huang who had Outstanding Papers 🏆

Mitigating racial bias from LLMs is a lot easier than removing it from humans! Can’t believe this happened at the best AI conference NeurIPS Conference We have ethical reviews for authors, but missed it for invited speakers? 😡

STATEMENT FROM TIKTOK: In agreement with our service providers, TikTok is in the process of restoring service. We thank President Trump for providing the necessary clarity and assurance to our service providers that they will face no penalties providing TikTok to over 170

🚀 Officially Dr. Zihao He! 🎉 I’m beyond excited to share that I have successfully defended my PhD thesis, “Aligning Large Language Models with Human Perspectives” at USC! This marks the culmination of an incredible journey—years of research, collaboration, and growth.

Check this work out! 🤗 We found that smoothed labels provided by a teacher model can mitigate the hallucinations from a student model! 🧐

Models can follow instructions—but can they speak like a community? STEER-BENCH shows that even top LLMs fall short on norm alignment across diverse online communities. Cultural steerability is still an open challenge. 👇

Glad to be part of this very promising project!

Seven years ago, I co-led a paper called 𝗛𝗼𝘁𝗽𝗼𝘁𝗤𝗔 that has motivated and facilitated many #AI #Agents research works since. Today, I'm asking that you stop using HotpotQA blindly for agents research in 2025 and beyond. In my new blog post, I revisit the brief history of

KV caching in LLMs, clearly explained (with visuals):

Finally!!!