🥳Excited to share VaPR at COLM 2025 (Montreal)! VaPR builds a preference dataset with hard-negative responses for LVLMs, reducing DPO overfitting to style/length cues and improving generalization. 📜 arxiv.org/abs/2510.01700 🌐 vap-r.github.io 🧵

Multilingual Routing in Mixture-of-Experts LLMs We present (1) an in-depth analysis of how MoE LLMs route multilingual texts, with very clear patterns + (2) a router intervention (steering) method that leads to consistent multilingual improvements! 🧵1/4 arxiv.org/pdf/2510.04694

New paper 📢 Most powerful vision-language (VL) reasoning datasets remain proprietary 🔒, hindering efforts to study their principles and develop similarly effective datasets in the open 🔓. Thus, we introduce HoneyBee, a 2.5M-example dataset created through careful data

.Kai-Wei Chang is getting a full house for his talk on “mathematical reasoning in visual context” at the Towards Comprehensive Reasoning in Vision-Language Models tutorial at #ICCV2025. Still time to come and engage in room 318A!

We’ve been running the UCLA NLP Seminar for a while now and realized it’s a waste not to share these amazing talks more broadly. So here’s our YouTube channel now! 🎥 Watch and subscribe to our channel for past and upcoming sessions: 👉 youtube.com/@uclanlp-plus #AI #UCLANLP

🤖💬AI agents can be easily persuaded (like Anthropic’s Claudius often giving discounts). 🤔Previous study on persuasion has been exclusively on text-only modality. We wonder: are AI agents more susceptible when presented with multimodal content? Introducing MMPersuade, a

Check out Yunzhi's work on circuit-aware editing for multi-hop knowledge!

Did Hinton win the Nobel Prize? 🤔

Extremely honored to be nominated for the UCLA Chancellor’s Award for Postdoctoral Research! Deeply grateful to Violet Peng and Kai-Wei Chang for the nomination, and to my friends at @UCLANLP for their support! I’m on the academic market now. Reach out if I might be a good fit!