8 papers
Mage-Flow: An Efficient Native-Resolution Foundation Model for Image Generation and Editing
Xinjie Zhang, Peng Zhang, Shicheng Zheng +21
Large-scale visual generators are increasingly capable but costly to train, fine-tune, and deploy. We introduce Mage-Flow, a compact 4B-scale generative stack for efficient text-to…
Demystifying Video Reasoning
Ruisi Wang, Zhongang Cai, Fanyi Pu +11
Recent advances in video generation have revealed an unexpected phenomenon: diffusion-based video models exhibit non-trivial reasoning capabilities. Prior work attributes this to a…
Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs
Trung Nguyen Quang, Yiming Gao, Fanyi Pu +3
When an omnimodal large language model accepts a question whose textual premise contradicts what it actually sees or hears, does the failure lie in perception or in action? Recent…
Scaling Spatial Intelligence with Multimodal Foundation Models
Zhongang Cai, Ruisi Wang, Chenyang Gu +26
Despite remarkable progress, multimodal foundation models still exhibit surprising deficiencies in spatial intelligence. In this work, we explore scaling up multimodal foundation m…
LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models
Kaichen Zhang, Bo Li, Peiyuan Zhang +8
The advances of large foundation models necessitate wide-coverage, low-cost, and zero-contamination benchmarks. Despite continuous exploration of language model evaluations, compre…
Memory-Efficient LLM Training by Various-Grained Low-Rank Projection of Gradients
Yezhen Wang, Zhouhao Yang, Brian K Chen +4
Building upon the success of low-rank adapter (LoRA), low-rank gradient projection (LoRP) has emerged as a promising solution for memory-efficient fine-tuning. However, existing Lo…