iiinova — where your digital self keeps evolving

Zhijing Jin @zhijingjin.bsky.social · 30/06/2025

🏛️Can we trust #LLMs to value democracy? Check our paper "Democratic or Authoritarian? Probing a New Dimension of Political Biases in LLMs". We find Llama4 has a stronger authoritarian bias, and all models show more bias in Mandarin than in English📜http://arxiv.org/abs/2506.12758 #AIBias #Democracy

050

Zhijing Jin @zhijingjin.bsky.social · 10/06/2025

📷 Check out our new position paper on how #LLM Agents can benefit frontier Theoretical Physics. We cover 🚀Current progress & key gaps ⚛️Potential for automating the Hypothesis-Verification cycle of scientific discovery 🔧Need for physics-specialized LLMs+tailored tool ecosystems

arxiv.org

Can Theoretical Physics Research Benefit from Language Agents?

Large Language Models (LLMs) are rapidly advancing across diverse domains, yet their application in theoretical physics research is not yet mature. This position paper argues that LLM agents can poten...

110

Zhijing Jin @zhijingjin.bsky.social · 28/05/2025

Pls Repost!📢 Unsure about NLP Ethics? Hoping to attend #ACL2025? How about applying for a virtual registration subsidy? All 3? We hear you!✨Tutorial: Navigating Ethical Challenges in NLP✨🗓️Sun, Jul 27, 14:00–17:30 CEST📍@ Hall M, Vienna + 🌍Online (hybrid!) ethics.aclweb.org/tutorials/AC...

ethics.aclweb.org

ACL 2025 Ethics Tutorial: Navigating Ethical Challenges in NLP - ACL Ethics

013

Zhijing Jin @zhijingjin.bsky.social · 14/04/2025

🌍 How do #LLMs handle trolley problems across cultures? We test them with 98K dilemmas in 107 languages, grounded in 40M+ human moral judgments. 💡 Spotlight @ICLR2025 (in Singapore✈️) | Best Paper #NeurIPS2024 Pluralistic Alignment Workshop 📄 Check out our paper: arxiv.org/abs/2407.02273 🧵👇

160

Zhijing Jin @zhijingjin.bsky.social · 11/04/2025

Honored to be selected for the Heidelberg Laureate Forum 2025! After attending Lindau Nobel Meeting (2024), grateful for another chance to learn from pioneers—#LINO24 for Physics Nobelists, now #HLF25 for CS/Math legends (Fields Medalists, Turing winners, ...). Really humbled!🙏

130

Zhijing Jin @zhijingjin.bsky.social · 08/04/2025

Check out my mentee's latest work on LLM Router attack! First work on this topic to the best of our knowledge. Read our paper at: zhijing-jin.com/files/papers... Great job to Aly Kassem!🎉 @uoftcompsci.bsky.social and MPI x.com/_AKassem/sta...

030

Zhijing Jin @zhijingjin.bsky.social · 08/04/2025

Check out our very popular "Tutorial on Causality for LLMs" at #NeurIPS2024: 2-hour detailed intro + a super insightful panel w/ @Yoshua_Bengio @bschoelkopf @Swarooprm7 @giambattista92 @tom4everitt on "The Role of Causality in Future LLMs" 📺 youtube.com/watch?v=8j1f... with slides

youtube.com

[NeurIPS 2024 Tutorial] Causality for Large Language Models

YouTube video by Zhijing Jin on AI Insights

070

Zhijing Jin @zhijingjin.bsky.social · 07/04/2025

Check out our latest work on LLM Coding Agents by more effective trajectory search! 🏆Beats GPT-4o & Claude 3.5 Sonnet in patch classification arxiv.org/pdf/2503.14269 🙌 Kudos to our great team: Vaibhav Aggarwal, Ojasv Kamal, Abhinav Japesh, Zhijing Jin, Bernhard Schölkopf. See x.com/kamalojasv18...

030

Zhijing Jin @zhijingjin.bsky.social · 07/04/2025

New faculty position opening in Germany with CISPA on AI and information security: jobs.cispa.saarland/jobs/detail/...

020