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。
DeepReinforce Team, Xiaoya Li, Xiaofei Sun +4 authors
GrandCode is a multi-agent reinforcement learning system that outperforms human competitors in competitive programming challenges by orchestrating specialized agent modules and employing novel reward policy optimization techniques.
50 篇论文 · 按点赞排序
Hongyuan Adam Lu, Z. L., Victor Wei +5 authors
A novel framework for improving large language model performance through textual frequency analysis, including laws, distillation, and curriculum training approaches.
Qihan Ren, Peng Wang, Ruikun Cai +8 authors
Supervised finetuning and reinforcement learning exhibit conditional cross-domain generalization in reasoning tasks, influenced by optimization dynamics, data quality, and model capability, with asymmetric outcomes between reasoning improvement and safety degradation.
Ziyu Ma, Shidong Yang, Yuxiang Ji +5 authors
SkillClaw enables collective skill evolution in multi-user LLM agent systems by aggregating user interactions to autonomously update and improve reusable skills across the ecosystem.
Yuxuan Zhang, Yubo Wang, Yipeng Zhu +18 authors
ClawBench presents a comprehensive evaluation framework with 153 real-world tasks across 144 platforms to test AI agents' ability to automate everyday online activities requiring complex multi-step workflows and document processing.
Jian Yang, Wei Zhang, Jiajun Wu +22 authors
Industrial software development lacks expert reasoning traces for hardware constraints, so a model was trained on error-driven reasoning chains and domain-specific execution traces to generate high-quality code reasoning and performance.
Chaoyou Fu, Haozhi Yuan, Yuhao Dong +16 authors
Video-MME-v2 presents a comprehensive benchmark for evaluating video understanding models through a progressive hierarchy and group-based evaluation to assess robustness and faithfulness.
DataFlow Team, Bohan Zeng, Daili Hua +38 authors
OpenWorldLib presents a standardized framework for advanced world models that integrate perception, interaction, and long-term memory capabilities for comprehensive world understanding and prediction.
Tencent Robotics X, HY Vision Team, Xumin Yu +19 authors
HY-Embodied-0.5 is a foundation model family for embodied agents featuring Mixture-of-Transformers architecture and iterative post-training for enhanced visual perception and reasoning capabilities.
Chenxu Yang, Chuanyu Qin, Qingyi Si +7 authors
RLSD combines reinforcement learning with verifiable rewards and self-distillation to achieve stable training with fine-grained updates and reliable policy direction from environmental feedback.
Bin Wang, Tianyao He, Linke Ouyang +40 authors
Training data engineering and optimized strategies improve document parsing performance without architectural changes, achieving state-of-the-art results on OmniDocBench v1.6.
Bowen Ye, Rang Li, Qibin Yang +10 authors
Claw-Eval addresses limitations in agent benchmarks by providing comprehensive evaluation across multiple modalities with trajectory-aware grading and safety assessments.
Weian Mao, Xi Lin, Wei Huang +5 authors
TriAttention addresses KV cache memory bottlenecks in LLMs by leveraging Q/K vector concentration in pre-RoPE space to improve key importance estimation and enable efficient long-context generation.
Zhengyang Sun, Yu Chen, Xin Zhou +4 authors
NUMINA enhances text-to-video diffusion models' numerical accuracy through a training-free identify-then-guide framework that uses attention heads to detect layout inconsistencies and refine object counts.
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.
Chanyoung Kim, Minwoo Kim, Minseok Kang +2 authors
Vision-Language-Action models show significant performance drops when handling paraphrased instructions due to surface-level matching rather than semantic understanding, highlighting the need for better linguistic generalization metrics.
Ailing Zeng, Casper Yang, Chauncey Ge +22 authors
A large-scale multimodal model for real-time conversational character performance generation that maintains identity consistency while enabling interactive, infinite-length video synthesis.
Lei Zhang, Junjiao Tian, Zhipeng Fan +9 authors
Process-driven image generation decomposes synthesis into iterative steps involving textual planning, visual drafting, textual reflection, and visual refinement, with step-wise supervision ensuring consistency and interpretability.
Yujiao Shen, Shulin Tian, Jingkang Yang +1 authors
A simple sliding-window approach using recent video frames outperforms complex memory-based streaming video understanding methods, revealing trade-offs between real-time perception and long-term memory capabilities.
Yuqi Zhou, Sunhao Dai, Changle Qu +3 authors
Retrieval models for agentic search should be trained directly from agent interaction data using a new paradigm that mines supervision from multi-step agent trajectories and incorporates relevance intensity through weighted optimization.
Zihan Wang, Chi Gui, Xing Jin +13 authors
Research identifies template collapse in multi-turn LLM agents as a hidden failure mode undetectable by entropy, proposing mutual information proxies and SNR-aware filtering to improve reasoning quality and task performance.
Jingyang Qiao, Weicheng Meng, Yu Cheng +6 authors
Memory Intelligence Agent framework integrates non-parametric and parametric memory systems with reinforcement learning to enable efficient reasoning and autonomous evolution in open-world environments.
Zigeng Chen, Gongfan Fang, Xinyin Ma +2 authors
DMax introduces a novel approach for efficient diffusion language models that reduces error accumulation during parallel decoding through self-refinement and unified training strategies.
Chenyu Zhou, Huacan Chai, Wenteng Chen +18 authors
Large language model agents are evolving beyond weight-based modifications to incorporate externalized components like memory, skills, and protocols that enhance reliability and coordination.
Hui Sun, Yun-Ji Zhang, Zheng Xie +4 authors
Researchers address the challenge of selecting correct code candidates from LLM-generated outputs by developing ACES, a method that ranks tests based on their ability to distinguish correct from incorrect code through leave-one-out evaluation and AUC consistency scoring.
Xudong Lu, Yang Bo, Jinpeng Chen +9 authors
AURA is an end-to-end streaming visual interaction framework that enables continuous video stream processing with real-time question answering and proactive responses through integrated context management and optimized deployment.
Wenbo Hu, Xin Chen, Yan Gao-Tian +3 authors
Gaussian GRPO addresses challenges in multimodal model training by using distributional matching to ensure gradient equity and stable reinforcement learning, enabling improved perception-reasoning balance in generalist models.
Tongbo Chen, Zhengxi Lu, Zhan Xu +13 authors
KnowU-Bench presents a comprehensive benchmark for personalized mobile agents that evaluates true preference inference and proactive assistance capabilities in real-world GUI environments.
Zhengqing Yuan, Hanchi Sun, Lichao Sun +1 authors
MegaTrain enables efficient training of large language models with over 100 billion parameters on a single GPU by utilizing host memory storage and optimized data streaming techniques.
Hyunsoo Cha, Wonjung Woo, Byungjun Kim +1 authors
Vanast is a unified framework that generates garment-transferred human animation videos by combining image-based virtual try-on and pose-driven animation in a single process, addressing issues like identity drift and garment distortion through triplet supervision and dual module architecture.
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