works on

From the 1 of 14 linked papers with an AI index.

activity
20242026
collaborators

14 papers

cs.CV2026

Reinforcing Egocentric Spatial Perception in Multimodal Large Language Models via Ego Scene Augmentation

Chi Kit Wong, Ye Pan, Yuanhuiyi Lyu +6

The paper proposes Ego Scene Augmentation (ESA), a framework that uses an Ego-element Graph to improve the spatial perception of multimodal large language models for egocentric vis…

cs.CV2026

Are We Using the Right Benchmark: An Evaluation Framework for Visual Token Compression Methods

Chenfei Liao, Wensong Wang, Zichen Wen +10

Recent efforts to accelerate inference in Multimodal Large Language Models (MLLMs) have largely focused on visual token compression. The effectiveness of these methods is commonly…

cs.CV2026

EgoIntent: A Pre-Outcome Micro-Step Benchmark for Understanding What, Why, and Next

Ye Pan, Chi Kit Wong, Yuanhuiyi Lyu +9

Egocentric video provides a natural modality for studying human behavior, but conventional visual understanding captures mainly observable scenes, objects, and actions rather than…

cs.CV2026

StruVis: Enhancing Reasoning-based Text-to-Image Generation via Thinking with Structured Vision

Yuanhuiyi Lyu, Kaiyu Lei, Ziqiao Weng +7

Reasoning-based text-to-image (T2I) generation requires models to interpret complex prompts accurately. Existing reasoning frameworks can be broadly categorized into two types: (1)…

cs.CV2025

Multimodal Spatial Reasoning in the Large Model Era: A Survey and Benchmarks

Xu Zheng, Zihao Dongfang, Lutao Jiang +17

Humans possess spatial reasoning abilities that enable them to understand spaces through multimodal observations, such as vision and sound. Large multimodal reasoning models extend…

cs.CV2025

Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generation

Yuanhuiyi Lyu, Chi Kit Wong, Chenfei Liao +5

Recent works have made notable advancements in enhancing unified models for text-to-image generation through the Chain-of-Thought (CoT). However, these reasoning methods separate t…