Explore · 每周精选
发现最受关注的研究论文,追踪研究趋势,订阅感兴趣的期刊与关键词。
50 篇论文 · 按点赞排序
Guanting Dong, Licheng Bao, Zhongyuan Wang +11 authors
AEPO, an agentic RL algorithm, addresses entropy-related challenges in web agent training, enhancing performance and stability across various datasets.
Chenghao Xiao, Hou Pong Chan, Hao Zhang +3 authors
Recent multimodal embedding approaches leveraging multimodal large language models (MLLMs) fine-tuned with contrastive learning (CL) have shown promising results, yet the underlying reasons behind their superiority remain underexplored. This work argues that a crucial advantage of MLLM-based approaches stems from implicit cross-modal alignment achieved during generative pretraining, where the language decoder learns to exploit multimodal signals within a shared representation space for generating unimodal outputs. Through analysis of anisotropy and kernel similarity structure, we empirically confirm that latent alignment emerges within MLLM representations, allowing CL to serve as a lightweight refinement stage. Leveraging this insight, we propose a Language-Centric Omnimodal Embedding framework, termed LCO-Emb. Extensive experiments across diverse backbones and benchmarks demonstrate its effectiveness, achieving state-of-the-art performance across modalities. Furthermore, we identify a Generation-Representation Scaling Law (GRSL), showing that the representational capabilities gained through contrastive refinement scales positively with the MLLM's generative capabilities. This suggests that improving generative abilities evolves as an effective paradigm for enhancing representation quality. We provide a theoretical explanation of GRSL, which formally links the MLLM's generative quality to the upper bound on its representation performance, and validate it on a challenging, low-resource visual-document retrieval task, showing that continual generative pretraining before CL can further enhance the potential of a model's embedding capabilities. Codes, models, and resources are available at https://github.com/LCO-Embedding/LCO-Embedding.
Tongyi DeepResearch Team, Baixuan Li, Bo Zhang +53 authors
Tongyi DeepResearch, a large language model with agentic capabilities, achieves top performance in various deep research tasks through an end-to-end training framework and automatic data synthesis.
Xiaoxi Li, Wenxiang Jiao, Jiarui Jin +8 authors
DeepAgent, an end-to-end deep reasoning agent, autonomously performs thinking, tool discovery, and action execution using memory folding and reinforcement learning, outperforming baselines in various tool-use and application tasks.
Mingyi Deng, Lijun Huang, Yani Fan +22 authors
InteractComp evaluates search agents' ability to recognize and resolve query ambiguity through interaction, revealing significant gaps in current models' capabilities.
Qiushi Sun, Jingyang Gong, Yang Liu +6 authors
A unified multimodal code corpus and model enable high-quality code generation from both text and visual inputs, outperforming commercial models.
Tianyu Fu, Zihan Min, Hanling Zhang +4 authors
Cache-to-Cache (C2C) enables direct semantic communication between LLMs using neural network projections, improving accuracy and reducing latency compared to text-based communication.
Justin Cui, Jie Wu, Ming Li +6 authors
A method is proposed to enhance long-horizon video generation by using sampled segments from self-generated long videos to guide student models, maintaining quality and consistency without additional supervision or retraining.
Enze Zhang, Jiaying Wang, Mengxi Xiao +7 authors
A new evaluation framework, DITING, and a reasoning-driven multi-agent evaluation framework, AgentEval, are introduced to assess the quality of web novel translations, revealing that Chinese-trained LLMs outperform larger foreign models.
Haoran Wei, Yaofeng Sun, Yukun Li
DeepSeek-OCR uses optical 2D mapping to compress long contexts, achieving high OCR precision with reduced vision tokens and demonstrating practical value in document processing.
Hanrong Ye, Chao-Han Huck Yang, Arushi Goel +29 authors
OmniVinci, an open-source omni-modal LLM, enhances cross-modal understanding and performance across audio, vision, and robotics applications with innovative architecture and efficient data curation.
Zichen Liu, Anya Sims, Keyu Duan +16 authors
GEM, an open-source environment simulator, facilitates experience-based learning for large language models by providing a standardized framework and diverse environments for training and benchmarking reinforcement learning algorithms.
Shijian Wang, Jiarui Jin, Xingjian Wang +6 authors
Video-Thinker, a multimodal large language model, autonomously reasons with videos using intrinsic grounding and captioning capabilities, achieving state-of-the-art performance on various video reasoning benchmarks.
Minjun Kim, Hyeonseok Lim, Hangyeol Yoo +10 authors
A large-scale investigation into constructing a fully open bilingual LLM for Korean using synthetic data demonstrates that such data can sustain pretraining and achieve performance comparable to multilingual baselines.
Zhiheng Xi, Xin Guo, Yang Nan +18 authors
BAlanced Policy Optimization with Adaptive Clipping (BAPO) addresses challenges in off-policy reinforcement learning by dynamically adjusting clipping bounds to improve sample efficiency, stability, and performance in large language models.
Hengyuan Xu, Wei Cheng, Peng Xing +8 authors
A diffusion-based model addresses copy-paste artifacts in text-to-image generation by using a large-scale paired dataset and a contrastive identity loss to balance identity fidelity and variation.
Zirui Guo, Xubin Ren, Lingrui Xu +2 authors
RAG-Anything is a unified framework that enhances multimodal knowledge retrieval by integrating cross-modal relationships and semantic matching, outperforming existing methods on complex benchmarks.
Runzhe Zhan, Yafu Li, Zhi Wang +5 authors
ExGRPO, a framework that prioritizes valuable reasoning experiences, improves and stabilizes reinforcement learning from verifiable rewards for large language models.
Luis Wiedmann, Orr Zohar, Amir Mahla +6 authors
FineVision, a large-scale and curated dataset, enhances vision-language models through rigorous data collection, de-duplication, and human oversight, leading to improved performance.
Jiahan Zhang, Muqing Jiang, Nanru Dai +14 authors
World-in-World evaluates generative world models in closed-loop environments, emphasizing task success over visual quality and revealing insights into controllability, data scaling, and compute allocation.
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