Xiao Wang
8 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
MirageBackdoor introduces a novel, highly stealthy backdoor attack that forces Large Language Models to generate correct reasoning steps (Think Well) but output an incorrect final answer (Answer Wrong), bypassing existing detection methods.
The paper argues that current search agents often verify existing knowledge rather than genuinely searching, and introduces LiveBrowseComp, a new benchmark to measure true evidence-driven discovery.
The paper introduces VIP-Net, a framework that leverages multi-modal spatio-temporal cues and a new dataset (Temporal-VIP) to accurately identify the most influential people in videos, overcoming the challenge of Temporal Importance Shift (TIS).
The paper argues that current search agents often verify existing knowledge rather than genuinely searching, and introduces LiveBrowseComp, a new benchmark to measure true evidence-driven discovery.
The paper introduces VIP-Net, a framework that leverages multi-modal spatio-temporal cues and a new dataset (Temporal-VIP) to accurately identify the most influential people in videos, overcoming the challenge of Temporal Importance Shift (TIS).
This paper introduces ARM, a discrete representation-based AutoRegressive Model for image understanding, generation, and editing.
This paper introduces PLAID-PRF, a method for performing Pseudo-Relevance Feedback (PRF) over PLAID, a centroid-based dense retrieval model, to improve retrieval effectiveness.
This paper proposes SPEAR, a spectral-disentangled MoE neural operator for large-scale PDE pre-training.
Papers
SPEAR: A Spectral-Disentangled MoE Neural Operator with Knowledge-Guided Expert Aggregation for Large-Scale PDE Pretraining
Dengdi Sun, Xiaoya Zhou, Xiao Wang, Wanli Lyu +2 more
This paper proposes SPEAR, a spectral-disentangled MoE neural operator for large-scale PDE pre-training.