Thrilled, grateful, and humbled to have won 3 outstanding paper awards at #EMNLP2024!!! Not even in my wildest dreams. Immense thanks to my amazing students and collaborators!
All three works are on evaluating LLMโ abilities in creative narrative generation. ๐งต๐
SPHERE has been accepted to ICLR 2026 and will be presented at Poster #214, Poster Session 4, Pavilion 3, 24 April, 3:15โ5:45 p.m.
Iโm unable to attend in person, our amazing advisor @VioletNPeng will help present this work! Stop by if youโre around.
๐ SteerMoE has been accepted to ICLR 2026 and will be presented at Poster #603, Poster Session 2, Pavilion 3, Thu 23 Apr, 3:15โ5:45 p.m.
๐ Iโm unable to attend in person, so many thanks to @LucasBandarkar for presenting on our behalf. Stop by if youโre around.
Introducing @NeoCognition, the agent lab for specialized intelligence.
Everyone needs experts, but human expertise does not scale.
Backed by $40M seed funding, we build self-learning agents that specialize across domains to make expertise abundant.
Excited to work with a world-class team (@sahaiamit, Raghu Meka,@VioletNPeng,@terrence_tao,@WeiWang1973) on AI for math discovery and thank @LaudeInstitute for the support. I'm hiring a postdoc in this direction. Please contact me by May 1, 2026, if youโre interested.
I'll be attending ICLR in a few weeks to present Multilingual Routing in Mixture-of-Experts
Lmk if you're attending and want to talk MoEs, interpretability, multilinguality, etc
Weโre excited to release TorchLean which is the first fully verified neural network framework in Lean. The Lean community has largely focused on pure mathematics. TorchLean expands this frontier toward verified neural network software and scientific computing. With the recent release of CSlib, we see this as another step toward a fully verified ML stack.
We support features:
1. Executable IEEE-754 floating-point semantics (and extensible alternative FP models) verified tensor abstractions with precise shape/indexing semantics
2. Formally verified autograd system for differentiation of NN programs Proof-checked certification / verification algorithms like CROWN (robustness, bounds, etc.)
3. PyTorch-inspired modeling API with eager-style development + export/lowering to a shared IR for execution and verification
Project page: https://t.co/YHpqhRbMQe
Paper: [2602.22631] TorchLean: Formalizing Neural Networks in Lean
Work done @Robertljg, Jennifer Cruden, Xiangru Zhong, @huan_zhang12 and @AnimaAnandkumar.
#MachineLearning #ScientificComputing #Lean
(2/4) Nanyun (Violet) Peng @VioletNPeng@UCLA is developing MAP-SE, a Multi-Agent Persuasion Simulation Engine that studies how influence emerges across adaptive agents with long-term memoryโmoving beyond simple one-on-one interactions. #AIResearch#AgenticAI
Extremely honored to be nominated for the @UCLA Chancellorโs Award for Postdoctoral Research!
Deeply grateful to @VioletNPeng and @kaiwei_chang for the nomination, and to my friends at @UCLANLP for their support!
Iโm on the academic market now. Reach out if I might be a good fit!
โจ Very overdue update:
I'll be starting as an Assistant Professor in CS at University of Minnesota, Twin Cities, Fall 2026. I will be recruiting PhD students!!
Please help me spread the word! [Thread] 1/n
๐ Excited to share that our ๐ ๐๐น๐๐ถ๐บ๐ผ๐ฑ๐ฎ๐น ๐๐๐น๐๐๐ฟ๐ฎ๐น ๐ฆ๐ฎ๐ณ๐ฒ๐๐ paper is accepted at TMLR 2025!
Grateful to all collaborators โ this continues our series of projects on ๐ฐ๐๐น๐๐๐ฟ๐ฎ๐น ๐๐ฒ๐ป๐๐ถ๐๐ถ๐๐ถ๐๐ ๐ฎ๐ป๐ฑ ๐๐ฎ๐ณ๐ฒ๐๐:
1๏ธโฃ SafeWorld: Geo-Diverse Safety Alignment (NeurIPS 2024)
2๏ธโฃ Evaluating Cultural & Social Awareness of LLM Web Agents (NAACL 2025)
3๏ธโฃ Multimodal Cultural Safety (TMLR 2025)
Looking forward to pushing this line of work further ๐โจ
Itโs a wrap! I hope yโall enjoyed #EMNLP25 as much as I did! Big shoutout to the photography team! All my photos in this post are taking from their websites. To all attendees: you should check out if you havenโt already!!