Yang Wang
5 papers · Latest:
When Are LLM Inferences Acceptable? User Reactions and Control Preferences for Inferred Personal Information
Users are more curious than distressed by LLM inferences about personal data, with acceptability depending on context and third-party use.
ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety
ProjLens demystifies MLLM backdoors, revealing how projector fine-tuning introduces vulnerabilities and detailing their low-rank structure and activation mechanism.
Closing the Observational Gap in Cosmic Dynamics: AI-Enabled Reconstruction of the Universe's Vorticity and Rotational Flow Morphology
AI reconstructs the universe's previously unobservable cosmic vorticity and rotational flows, reinforcing the concordance cosmological model.
VAG: Dual-Stream Video-Action Generation for Embodied Data Synthesis
VAG is a dual-stream flow-matching framework that jointly generates aligned video and action pairs for embodied data synthesis, improving robot policy generalization.
ViVa: A Video-Generative Value Model for Robot Reinforcement Learning
ViVa is a video-generative value model that improves robot reinforcement learning by using a pretrained video generator to estimate future dynamics and task value.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.