Proposed BALSa framework uses synthetic data to improve audio-language alignment, reducing hallucinations and enhancing reasoning abilities.

Chun-Yi Kuan, Hung-yi Lee

HunyuanVideo-Avatar uses a multimodal diffusion transformer to generate high-fidelity audio-driven animations for multiple characters.

Yi Chen, Sen Liang, Zixiang Zhou et al.

Proposes GenICL, using LLM feedback for direct demonstration selection optimization, significantly improving ICL performance.

Zheng Zhang, Shaocheng Lan, Lei Song et al.

Proposes density-ratio-free doubly robust kernel estimators for proxy causal learning, effective in high-dimensional continuous treatments.

Bariscan Bozkurt, Houssam Zenati, Dimitri Meunier et al.

Proposes a dynamic multi-agent orchestration framework using reinforcement learning, improving task performance by 15% and reducing tokens by 30%.

Yufan Dang, Chen Qian, Xueheng Luo et al.

DiffVLA integrates vision-language models with hybrid sparse-dense diffusion for autonomous driving, achieving 45.0 PDMS on navsim-v2.

Anqing Jiang, Yu Gao, Zhigang Sun et al.

DriveX employs Omni Scene Modeling to learn generalizable scene dynamics, reducing Chamfer distance by 6% over 3s prediction, and excels in multi-task autonomous driving benchmarks.

Chen Shi, Shaoshuai Shi, Kehua Sheng et al.

CreatiDesign integrates multi-conditional diffusion transformer with multimodal attention masks, achieving precise control over heterogeneous design elements.

Hui Zhang, Dexiang Hong, Maoke Yang et al.

Proposes LatentR3, a reinforcement learning framework for efficient, explicit-chain-free recommendation with 17% performance gain.

Yang Zhang, Wenxin Xu, Xiaoyan Zhao et al.

Self-distillation framework for flow map learning accelerates consistency models, achieving 10-100× inference speedup with high stability.

Nicholas M. Boffi, Michael S. Albergo, Eric Vanden-Eijnden

This survey reviews 26 state-of-the-art Reasoning Segmentation (RS) models, emphasizing multi-stage reasoning, knowledge integration, and pixel-level mask generation.

Yiqing Shen, Chenjia Li, Fei Xiong et al.

GainRAG aligns retriever and LLM preferences via the 'gain' metric, boosting performance across six datasets.

Yi Jiang, Sendong Zhao, Jianbo Li et al.

CoT-RVS leverages zero-shot chain-of-thought reasoning with pretrained multimodal models for video segmentation, outperforming previous methods by over 10% in J&F metrics.

Shiu-hong Kao, Yu-Wing Tai, Chi-Keung Tang

Reinforcement fine-tuning enhances reasoning in multimodal LLMs, significantly boosting OpenAI-o1 and DeepSeek-R1 performance.

Haoyuan Sun, Jiaqi Wu, Bo Xia et al.

TriSense introduces a multimodal LLM with adaptive query-based fusion, trained on TriSense-2M with 2 million long videos, achieving state-of-the-art results in video understanding.

Zinuo Li, Xian Zhang, Yongxin Guo et al.

CapBencher reduces Bayes accuracy by randomizing answers to detect LLM test-set overfitting.

Takashi Ishida, Thanawat Lodkaew, Ikko Yamane

AlignXplore model infers personalized preferences from behavioral signals, achieving a 15.49% performance boost.

Jia-Nan Li, Jian Guan, Wei Wu et al.

Introduces Semi-Simplicial Neural Networks (SSNs) for directed higher-order relation modeling, achieving state-of-the-art brain activity decoding accuracy.

Manuel Lecha, Andrea Cavallo, Francesca Dominici et al.

Trinity-RFT offers a unified, scalable framework for reinforcement fine-tuning of large language models, supporting multi-mode, multi-device training with high efficiency.

Xuchen Pan, Yanxi Chen, Yushuo Chen et al.

TextFlux, a DiT-based multilingual scene text synthesis model, achieves high fidelity without OCR, trained with only 1% data of competing methods.

Yu Xie, Jielei Zhang, Pengyu Chen et al.