← Search

Haihong Hao

2 accepted papers

2026

GeoSense: Internalizing Geometric Necessity Perception for Multimodal Reasoning

ICML 2026poster

Advancing towards artificial superintelligence requires rich and intelligent perceptual capabilities. A critical frontier in this pursuit is overcoming the limited spatial understanding of Multimodal Large Language Models (MLLMs), where geometry information is essential. Existing methods often addre…

Cited by 0Source ↗Scholar ↗
2026

Progressive Online Video Understanding with Evidence-Aligned Timing and Transparent Decisions

ICLR 2026poster

Visual agents operating in the wild must respond to queries precisely when sufficient evidence first appears in a video stream, a critical capability that is overlooked by conventional video LLMs evaluated in offline settings. The shift to an online, streaming paradigm introduces significant challen…

Cited by 0Source ↗Scholar ↗