8 papers · 1 filter
EventPrune: Cascaded Event-Assisted Token Pruning for Efficient First-Person Dynamic Spatial Reasoning
Pengtao Ma, Ziliang Zhou, Ciyu Ruan +7
First-person dynamic spatial reasoning requires models to track continuous motion and precise geometric structure, but the quadratic attention cost of Transformer-based Video-LLMs…
Progressive Supernet Training for Efficient Visual Autoregressive Modeling
Xiaoyue Chen, Yuling Shi, Kaiyuan Li +5
Visual Auto-Regressive (VAR) models significantly reduce inference steps through the "next-scale" prediction paradigm. However, progressive multi-scale generation incurs substantia…
EgoPrune: Efficient Token Pruning for Egomotion Video Reasoning in Embodied Agent
Jiaao Li, Kaiyuan Li, Chen Gao +2
Egomotion videos are first-person recordings where the view changes continuously due to the agent's movement. As they serve as the primary visual input for embodied AI agents, maki…
DIMM: Decoupled Multi-hierarchy Kalman Filter for 3D Object Tracking
Jirong Zha, Yuxuan Fan, Kai Li +4
State estimation is challenging for 3D object tracking with high maneuverability, as the target's state transition function changes rapidly, irregularly, and is unknown to the esti…
Balanced Token Pruning: Accelerating Vision Language Models Beyond Local Optimization
Kaiyuan Li, Xiaoyue Chen, Chen Gao +2
Large Vision-Language Models (LVLMs) have shown impressive performance across multi-modal tasks by encoding images into thousands of tokens. However, the large number of image toke…
The Point, the Vision and the Text: Does Point Cloud Boost Spatial Reasoning of Large Language Models? A Bias-Controlled Study
Weichen Zhang, Ruiying Peng, Xin Zeng +9
3D Large Language Models (LLMs) leveraging spatial information in point clouds for 3D spatial reasoning attract great attention. Despite some promising results, the advantages of p…
PLPHP: Per-Layer Per-Head Vision Token Pruning for Efficient Large Vision-Language Models
Yu Meng, Kaiyuan Li, Chenran Huang +4
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across a range of multimodal tasks. However, their inference efficiency is constrained by the large n…
Understanding and Evaluating Hallucinations in 3D Visual Language Models
Ruiying Peng, Kaiyuan Li, Weichen Zhang +3
Recently, 3D-LLMs, which combine point-cloud encoders with large models, have been proposed to tackle complex tasks in embodied intelligence and scene understanding. In addition to…