Multimedia

Aditi, Niket Agarwal, Arslan Ali et al.

This work provides a scalable, general-purpose backbone for embodied agents by unifying multiple modalities into a single framework, which is a significant step for Physical AI research.

Zitong Xu, Huiyu Duan, Zhongpeng Ji et al.

This work addresses a bottleneck in scalable human feedback for image editing, enabling better evaluation and optimization of editing models, though it is incremental in building upon existing MLLM and reinforcement learning techniques.

Jizhan Fang, Buqiang Xu, Zhixian Wang et al.

For LLM agents operating in dynamic environments, FluxMem addresses the brittleness of static memory by enabling adaptive connectivity evolution, leading to consistent SOTA results across diverse benchmarks.

Junchao Liao, Zhenghao Zhang, Xiangyu Meng et al.

This addresses the challenge of physical coherence in audio-video generation for applications like media production, though it is incremental as it builds on existing methods with a novel kinematic prior.

Guikun Chen, Wenguan Wang

It addresses the need for a comprehensive overview of this emerging method for researchers in computer graphics and vision, but it is incremental as it surveys existing developments rather than introducing new findings.

Quanhao Li, Zhen Xing, Rui Wang et al.

This work addresses efficiency and quality issues in video generation for applications requiring precise motion control, though it is incremental as it builds on existing adapter and distillation techniques.

Yi Su, Jisheng Bai, Qisheng Xu et al.

This is an incremental survey that helps researchers and practitioners in audio-centric AI by summarizing existing technologies and providing references for practical applications.

Zhiqiu Lin, Chancharik Mitra, Siyuan Cen et al.

For researchers and practitioners in video understanding and generation, this work provides a scalable method to produce high-quality, structured captions that improve both VLM performance and text-to-video generation control.

Zhifei Xie, Zihang Liu, Ze An et al.

This work addresses the need for a single model that can handle multiple streaming audio tasks (e.g., voice chatting, ASR) in real time, unifying capabilities that were previously separate.

Jiahao Meng, Yue Tan, Qi Xu et al.

For researchers in video understanding and MLLMs, this survey offers a structured framework to categorize and compare approaches, but it is an incremental synthesis of existing work rather than a novel contribution.

Matan Ben-Yosef, Tavi Halperin, Naomi Ken Korem et al.

This addresses the need for modular and efficient control in audio-visual generation for researchers and practitioners, offering a significant improvement over monolithic or costly methods.

Shuzhao Xie, Junchen Ge, Weixiang Zhang et al.

For practitioners deploying 3DGS in storage-constrained environments, this method provides a size-aware codec that accurately meets target budgets without retraining.

Ailing Zeng, Casper Yang, Chauncey Ge et al.

This addresses the problem of creating lifelike virtual characters for conversational agents, live streaming, and games, representing a novel method rather than an incremental improvement.

Yuheng Chen, Qingdong He, Teng Hu et al.

This work addresses the underexplored problem of multimodal customization for simultaneous identity preservation in audio-video generation, offering a solution for content creators needing consistent character voices and appearances.

Yifan Shen, Jiateng Liu, Xinzhuo Li et al.

This work addresses the problem of simulating dynamic egocentric environments for applications like smart-glasses, though it is incremental as it builds on existing generative world models.

Yale Song, Yiwen Song, Nick Losier et al.

For AI video generation, Co-Director addresses semantic drift in agentic pipelines, offering a principled optimization approach that generalizes to cinematic narratives.

Ziyi He, Yushi Feng, Shuangyu Yang et al.

This work addresses the safety-critical need for better multimodal AI in dental clinical routing, though it is incremental as it introduces a new benchmark without a novel method.

Rui Yang Tan, Yujia Hu, Roy Ka-Wei Lee

This addresses a safety vulnerability in MLLMs for users relying on visual reasoning, though it is incremental as it builds on existing jailbreak benchmarks.

Yaofeng Su, Yuming Li, Zeyue Xue et al.

This enables real-time applications for joint audio-visual generation, addressing a bottleneck in multi-modal AI systems, though it is incremental as it builds on existing diffusion models.

Yuzhe Weng, Haotian Wang, Yuanhong Yu et al.

This work addresses the problem of generating realistic and controllable talking head videos for applications like virtual avatars and video editing, representing a novel paradigm rather than an incremental improvement.