12 papers
CoEvoWhen: Policy-Tool Coevolution for Ultra-Long Video Temporal Grounding
Yiduo Jia, Muzhi Zhu, Jinchuan Shi +4
Ultra-long video temporal grounding requires balancing long-range evidence search with fine-grained event understanding under a limited visual budget, yet existing agentic methods…
Metric-Bench: Exploring In-context Spatial Metric Reasoning in VLMs for Indoor Scenes
Yuling Xi, Haokai Zhang, Muzhi Zhu +10
Metric reasoning is a critical and challenging task for Vision Language Models (VLMs), playing a pivotal role in embodied AI tasks such as robotic manipulation and autonomous navig…
CoMem: Reusing Transformer Depth across Queries with Persistent Intermediate Residuals
Hanzuo Liu, Xuan Qi, Chunyu Liu +6
Repeated queries over shared documents repeatedly execute the same lower transformer layers. We introduce CoMem, which makes split depth j an explicit reusable-context axis: write…
Exploring Spatial Intelligence from a Generative Perspective
Muzhi Zhu, Shunyao Jiang, Huanyi Zheng +9
Spatial intelligence is essential for multimodal large language models, yet current benchmarks largely assess it only from an understanding perspective. We ask whether modern gener…
LLaDA2.1: Speeding Up Text Diffusion via Token Editing
Tiwei Bie, Maosong Cao, Xiang Cao +47
While LLaDA2.0 showcased the scaling potential of 100B-level block-diffusion models and their inherent parallelization, the delicate equilibrium between decoding speed and generati…
SE360: Semantic Edit in 360 Panoramas via Hierarchical Data Construction
Haoyi Zhong, Fang-Lue Zhang, Andrew Chalmers +1
While instruction-based image editing is emerging, extending it to 360 panoramas introduces additional challenges. Existing methods often produce implausible results in bot…
Preserving Source Video Realism: High-Fidelity Face Swapping for Cinematic Quality
Zekai Luo, Zongze Du, Zhouhang Zhu +7
Video face swapping is crucial in film and entertainment production, where achieving high fidelity and temporal consistency over long and complex video sequences remains a signific…
NoTVLA: Semantics-Preserving Robot Adaptation via Narrative Action Interfaces
Zheng Huang, Mingyu Liu, Xiaoyi Lin +9
Vision-Language-Action (VLA) models represent a pivotal advance in embodied intelligence, yet they confront critical barriers to real-world deployment, most notably catastrophic fo…
GUI-Shepherd: Reliable Process Reward and Verification for Long-Sequence GUI Tasks
Cong Chen, Kaixiang Ji, Hao Zhong +9
Autonomous agents for long-sequence Graphical User Interface tasks are hindered by sparse rewards and the intractable credit assignment problem. To address these challenges, we int…
Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System Collaboration
Hao Zhong, Muzhi Zhu, Zongze Du +6
Long-horizon video-audio reasoning and fine-grained pixel understanding impose conflicting requirements on omnimodal models: dense temporal coverage demands many low-resolution fra…
SafeMove-RL: A Certifiable Reinforcement Learning Framework for Dynamic Motion Constraints in Trajectory Planning
Tengfei Liu, Haoyang Zhong, Jiazheng Hu +1
This study presents a dynamic safety margin-based reinforcement learning framework for local motion planning in dynamic and uncertain environments. The proposed planner integrates…
ACTIVE-o3: Empowering MLLMs with Active Perception via Pure Reinforcement Learning
Muzhi Zhu, Hao Zhong, Canyu Zhao +9
Active vision, also known as active perception, refers to actively selecting where and how to look in order to gather task-relevant information. It is a critical component of effic…