This is a cool paper! arxiv.org/abs/2503.24187

Join me and make a gift to W&M on One Tribe One Day! #OneTribeOneDay givecampus.com/schools/Willia…

Welcome to submit papers/contribute/pay attention to the workshop on M3RD: Multi-Modal Mathematical Reasoning in Documents! nlpr.ia.ac.cn/pal/Activities…

I will be at #ICLR2025 ICLR 2026 during April 26 to 28. Welcome to connect for ML, AI, LLM, alignment, social science, and foundation models:)

Our demo paper of Value Compass Benchmarks has been accepted by ACL 2025! 🥳 Want to know your model's underlying value orientations? Just check it! Thanks for all collaborators! Jindong Wang Paper link: lnkd.in/gPjGYuTn Online Benchmark: lnkd.in/gH7sCz22

We are organizing the workshop on Federated Learning with Generative AI at IJCAIconf 2025. Please consider submitting your papers:) federated-learning.org/FedGenAI-ijcai…

Thanks for highlighting:)

Welcome to our tutorial at #CVPR2026 on June 11 about large multi-modal model evaluation led by Kaijie Zhu and UCSB students. We also have a paper SoftVQ-VAE. lmm-understand.github.io

Hitting 20000 citations:) Thank you for finding my research useful.

I will be at #ICML in Vancouver during July 14-17. Connections are welcome!

We are organizing an International Programme on AI Evaluation and will recruit one Program Manager: existence.org/jobs. Please apply if you are interested!

Our LLM evaluation survey is the No.1 most downloaded paper and No.4 most cited paper in ACM TIST :) Evaluation truly becomes one of the hottest topics today! arxiv.org/abs/2307.03109

The mediocre #GPT5 means our research in academic labs can still be alive🤣 Thank you OpenAI for not killing our jobs.

In the Era of Large Models, What is the Research Ceiling for Ordinary People? Sharing my recent blog article: jd92wang.notion.site/In-the-Era-of-…

Today is the first class of the "Generative AI" course in this fall semester. The students are energetic. Will try my best to make the course worth it:)

Busy with your ICLR submission? Sharing some tips for improving your LaTex paper writing: jd92wang.notion.site/Tips-for-impro…

New reviewers needed! forms.gle/unzDQecPz3smeK…

Big news! We’re opening doors to the International Programme on AI Evaluation: Capabilities & Safety, an initiative co-developed with colleagues from Cambridge, Microsoft Research, the EU AI Office, Deep Mind, and others. Over four months (online + one week in Valencia), we’ll

How to sufficiently survey a research topic? Some thoughts: jd92wang.notion.site/How-to-suffici…

My take: The new paper making the rounds is basically a worse version of the ADeLe paper by JH Orallo and friends. kinds-of-intelligence-cfi.github.io/ADELE/