10 papers
Revitalizing the Beginning: Avoiding Storage Dependency for Model Merging in Continual Learning
Xi Wang, Cheng Deng
Model merging provides a compelling paradigm for integrating specialized expertise into a unified multi-task model, a goal that aligns naturally with the sequential knowledge acqui…
VisPCO: Visual Token Pruning Configuration Optimization via Budget-Aware Pareto-Frontier Learning for Vision-Language Models
Huawei Ji, Yuanhao Sun, Yuan Jin +4
Visual token pruning methods effectively mitigate the quadratic computational growth caused by processing high-resolution images and video frames in vision-language models (VLMs).…
SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs
Siting Wang, Minnan Pei, Luoyang Sun +6
Humans can imagine and manipulate visual images mentally, a capability known as spatial visualization. While many multi-modal benchmarks assess reasoning on visible visual informat…
-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs
Siting Wang, Xiaofeng Wang, Zheng Zhu +7
Flow-based vision-language-action (VLA) models excel in embodied control but suffer from intractable likelihoods during multi-step sampling, hindering online reinforcement learning…
Hardware Co-Design Scaling Laws via Roofline Modelling for On-Device LLMs
Luoyang Sun, Jiwen Jiang, Yifeng Ding +9
Vision-Language-Action Models (VLAs) have emerged as a key paradigm of Physical AI and are increasingly deployed in autonomous vehicles, robots, and smart spaces. In these resource…
LoopServe: An Adaptive Dual-phase LLM Inference Acceleration System for Multi-Turn Dialogues
Haoyang Li, Zhanchao Xu, Yiming Li +9
Multi-turn dialogues are essential in many real-world applications of large language models, such as chatbots and virtual assistants. As conversation histories become longer, exist…