Trends · 研究趋势
数据来自 Hugging Face 论文的 AI 提取关键词,按月统计研究方向的增长与热度。
Boqiang Zhang, Lei Ke, Ruihan Yang +5 authors
Penguin-VL demonstrates that text-only initialized vision encoders can achieve superior performance in multimodal understanding tasks compared to traditional contrastive pretraining methods, enabling efficient deployment on resource-constrained devices.
Jiabin Hua, Hengyuan Xu, Aojie Li +4 authors
A diffusion framework called PixelSmile is introduced that disentangles facial expression semantics through symmetric joint training and contrastive learning to enable precise, controllable, and fine-grained expression editing with robust identity preservation.
Junyao Gao, Sibo Liu, Jiaxing Li +6 authors
MegaStyle presents a scalable data curation pipeline for creating high-quality, style-consistent datasets using large generative models and proposes style-supervised contrastive learning for effective style representation extraction.
Jiuhai Chen, Jianwei Yang, Haiping Wu +4 authors
Florence-VL enhances multimodal language models with improved visual features from Florence-2, achieving state-of-the-art results across various benchmarks through depth-breath fusion and a novel training recipe.
Corby Rosset, Ching-An Cheng, Arindam Mitra +3 authors
A Direct Nash Optimization (DNO) algorithm is introduced for post-training large language models using preference feedback, achieving state-of-the-art performance with polynomial parameters.
Weiquan Huang, Aoqi Wu, Yifan Yang +8 authors
LLM2CLIP leverages large language models to enhance CLIP's multimodal representation learning, enabling better handling of complex image captions by fine-tuning the LLM and using contrastive learning.
Yi Wang, Yinan He, Yizhuo Li +10 authors
InternVid, a large-scale video-text multimodal dataset, and ViCLIP, a video-text representation model based on ViT-L, enable advanced video and text understanding and generation tasks.
Tingyu Song, Mingxin Li, Yanzhao Zhang +5 authors
CORE distills compositional ranking judgments from a cross-attentive reranker into an embedding model via synthesized multi-level candidates and a Rank-KL objective, improving compositional retrieval without degrading standard performance.
Yue Wang, Hung Le, Akhilesh Deepak Gotmare +3 authors
CodeT5+, a family of flexible encoder-decoder LLMs initialized with frozen LLMs and trained with a mixture of pretraining objectives, achieves state-of-the-art performance across various code-related tasks.
Paul S. Scotti, Atmadeep Banerjee, Jimmie Goode +9 authors
MindEye, a novel fMRI-to-image approach, retrieves and reconstructs viewed images with high accuracy using specialized submodules and diffusion prior in latent space.
北京市昌平区探索星信息技术及软件开发工作室
京ICP备2026059466号