6 papers
MultiBreak: A Scalable and Diverse Multi-turn Jailbreak Benchmark for Evaluating LLM Safety
Jialin Song, Xiaodong Liu, Weiwei Yang +4
We present MultiBreak, a scalable and diverse multi-turn jailbreak benchmark to evaluate large language model (LLM) safety. Multi-turn jailbreaks mimic natural conversational setti…
FluidGaussian: Propagating Simulation-Based Uncertainty Toward Functionally-Intelligent 3D Reconstruction
Yuqiu Liu, Jialin Song, Marissa Ramirez de Chanlatte +5
Real objects that inhabit the physical world follow physical laws and thus behave plausibly during interaction with other physical objects. However, current methods that perform 3D…
Lean Finder: Semantic Search for Mathlib That Understands User Intents
Jialin Lu, Kye Emond, Kaiyu Yang +3
We present Lean Finder, a semantic search engine for Lean and mathlib that understands and aligns with the intents of mathematicians. Progress in formal theorem proving is often hi…
WildSmoke: Ready-to-Use Dynamic 3D Smoke Assets from a Single Video in the Wild
Yuqiu Liu, Jialin Song, Manolis Savva +1
We propose a pipeline to extract and reconstruct dynamic 3D smoke assets from a single in-the-wild video, and further integrate interactive simulation for smoke design and editing.…
PDE-Controller: LLMs for Autoformalization and Reasoning of PDEs
Mauricio Soroco, Jialin Song, Mengzhou Xia +3
While recent AI-for-math has made strides in pure mathematics, areas of applied mathematics, particularly PDEs, remain underexplored despite their significant real-world applicatio…
Data-Efficient Operator Learning via Unsupervised Pretraining and In-Context Learning
Wuyang Chen, Jialin Song, Pu Ren +3
Recent years have witnessed the promise of coupling machine learning methods and physical domain-specific insights for solving scientific problems based on partial differential equ…