FlowVLA introduces explicit motion reasoning via Visual CoT, significantly improving long-term visual prediction and robot policy efficiency.

Zhide Zhong, Haodong Yan, Junfeng Li et al.

Proposes HLLM-Creator, a hierarchical LLM framework for efficient personalized content generation, achieving 0.476% CTR increase in industrial ad scenarios.

Junyi Chen, Lu Chi, Siliang Xu et al.

Pandora uses Python Pandas for unified structured knowledge reasoning, enhancing cross-task performance.

Yongrui Chen, Junhao He, Linbo Fu et al.

JCo-MVTON leverages multi-modal diffusion transformers for mask-free virtual try-on, achieving SOTA performance on DressCode dataset.

Aowen Wang, Wei Li, Hao Luo et al.

Proposes UQ, a dynamic benchmark assessing LLMs on 500 unsolved questions sourced from Stack Exchange, integrating validator-assisted screening and community verification.

Fan Nie, Ken Ziyu Liu, Zihao Wang et al.

Music generation evaluation lacks standardization; proposes detailed taxonomy and future research directions.

Faria Binte Kader, Santu Karmaker

CMASE integrates generative agents with ethnography, enabling real-time human-in-the-loop social simulation and intervention.

Hanzhong Zhang, Muhua Huang, Jindong Wang

UniGen framework uses CoMoE module and WeaveNet mechanism for efficient controllable image generation, excelling on Subjects-200K and MultiGen-20M datasets.

Guoqing Zhang, Xingtong Ge, Lu Shi et al.

CORE-RAG employs performance-driven reinforcement learning to compress context to 3%, boosting EM scores by 3.3 points, outperforming full-document baseline.

Ziqiang Cui, Yunpeng Weng, Xing Tang et al.

PowerChain automates DG analysis with verifiable DAG agents, reaching up to ~144% improvement.

Emmanuel O. Badmus, Peng Sang, Dimitrios Stamoulis et al.

PREMIR framework leverages cross-modal question generation with large language models to achieve state-of-the-art zero-shot multimodal document retrieval, outperforming baselines.

Yejin Choi, Jaewoo Park, Janghan Yoon et al.

Proposes entity-oriented search with graph query language, achieving state-of-the-art results on WikiTQ and TabFact datasets.

Thi-Nhung Nguyen, Hoang Ngo, Dinh Phung et al.

LHM-Humanoid achieves long-horizon human motion control for continuous object transport in cluttered scenes, outperforming baselines significantly.

Haozhuo Zhang, Jingkai Sun, Michele Caprio et al.

MV-RAG integrates retrieval with multiview diffusion, significantly improving out-of-domain concept 3D generation with 15%+ performance boost.

Yosef Dayani, Omer Benishu, Sagie Benaim

Spectral decomposition of interpolation spaces [L2(ν), H]θ,r extends spectral analysis from r=2 to r∈[1,∞], enabling Banach space representation and embedding criteria.

Michael Bitzer, Ingo Steinwart

OPERA combines reinforcement learning with a hierarchical planner-executor architecture, achieving 57.3% EM on HotpotQA and superior multi-hop reasoning.

Yu Liu, Yanbing Liu, Fangfang Yuan et al.

GelSLAM employs surface normals and curvature for real-time, high-precision tactile SLAM, achieving submillimeter accuracy over long trajectories.

Hung-Jui Huang, Mohammad Amin Mirzaee, Michael Kaess et al.

HRES numerical model outperforms AI models in forecasting record-breaking weather extremes.

Zhongwei Zhang, Erich Fischer, Jakob Zscheischler et al.

StreamMem employs query-agnostic KV cache compression, enabling efficient long video understanding with fixed memory, outperforming state-of-the-art methods.

Yanlai Yang, Zhuokai Zhao, Satya Narayan Shukla et al.

Proposes FRISBI, an inductive SBI framework combining mini-batch OT and conditional normalizing flows, addressing model misspecification.

Ortal Senouf, Antoine Wehenkel, Cédric Vincent-Cuaz et al.