Joint audio-video generation models have made substantial progress in visual quality and audio-visual synchronization. However, they still provide limited control over when shot transitions occur and dialogue is spoken. This limitation constrains their application in script-driven content creation, where timing errors...
Yi-Chen Liu, Quan-Wei Zhang, Hao-Zhe Wang et al.· 0 citations
OmniPack is proposed, a training-free framework that coordinates structural compression before the LLM with task-relevant semantic refinement within the LLM, and consistently achieves the best performance-efficiency trade-off across diverse retention ratios, outperforming all existing methods.
Wan-Shun Su, Yang Shi, Fei Liu et al.· 2 citations
The Thinking Reward Model (TRM), a paradigm that explicitly determines what matters for each case before judging how well the candidate performs, is proposed, which formulates case-adaptive rubrics, performs rubric-guided assessment, and produces fine-grained pointwise rewards.
Xue-Yuan Bai, Zhen-Chen Tang, Yang Shi et al.· 0 citations
This work introduces SPACE (Subject-centric Poses, Appearance, and Characteristics Evaluation), a benchmark designed to evaluate subject-centered spatial understanding, and designs a rubric-based reward for Group Relative Policy Optimization that explicitly penalizes structurally critical spatial errors during alignmen...
Bozhou Li, Jia-Hang Zhang, Yue Ding et al.· 0 citations
This work introduces HiRAE (Hierarchical Representation Autoencoder), which learns a hierarchical fusion framework over the full encoder hierarchy to improve reconstruction fidelity while maintaining compatibility with generative modeling.
Xuan-Yu Zhu, Yan Bai, Yang Shi et al.· 0 citations
NovGauge is presented, a human-anchored benchmark for fine-grained novelty assessment diagnosis, and a cascading diagnostic pipeline that verifies per-dimension correctness, evidence grounding, and logical support is proposed.
Guo-Qiang Zhang, Ke-Xin Tan, Ming Zhang et al.· 0 citations
Talking-video character replacement requires coordinated transfer of appearance and voice while preserving the source motion, scene, linguistic content, and audio-video timing. Existing methods use separately optimized models for the two modalities, making audio-visual consistency difficult to enforce. We present UniSw...
Yuxuan Zhang, H. Xiong, Jiayi Song et al.· 0 citations
Human evaluation confirms that RefCaptioner achieves the best overall performance among the open-source models while remaining competitive on standard video captioning benchmarks, and human evaluation further confirms that its captions are preferred by annotators and enable more source-faithful video reconstruction wit...
Teng-Fei Liu, Yang Shi, Yu-Ran Wang et al.· arXiv.org· 0 citations
The fundamental goal of agentic visual reasoning is to improve the success rate of multimodal large language models (MLLMs) on complex tasks, rather than merely equipping them with a sophisticated yet inefficient reasoning paradigm. In this work, we rethink agentic visual reasoning through two key dimensions of tool us...
Qixun Wang, Yang Shi, Le-Tian Cheng et al.· arXiv.org· 0 citations
Positive--Direction Matching (PDM), a branch-aware OPD objective that separately constrains the positive prediction and the CFG conditional direction, is introduced to dense-to-sparse video control, where naive guided matching is highly sensitive to inference guidance scales, while branch-aware supervision enables more...
Bingnan Li, Hao-Zhe Wang, H. Xiong et al.· arXiv.org· 2 citations
MultiRef-Compass is introduced, a unified benchmark for MR2AV generation that integrates automatic metrics with a rejudging-enhanced MLLM-as-a-Judge framework, enabling scalable and auditable evaluation of both perceptual fidelity and reference-conditioned composition.
GeoLens is developed, a multi-tool visual reasoning MLLM for UHR RS, and a tool-attention-focused reinforcement learning algorithm that concentrates optimization on critical tool-use decisions, including when to invoke tools, which tool to select, where to apply it, and how to interpret tool outputs is proposed.
