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。
Junxiang Xu, Ruisi Wang, Fanyi Pu +49 authors
VBBR-Pro introduces a closed-loop testbed that enables scalable, verifiable, and controllable native visual reasoning through generation across diverse visual substrates.
50 篇论文 · 按点赞排序
Apodex Team, B. An, B. Li +68 authors
Apodex 1.1 improves sustained, verifiable progress on complex real-world tasks by scaling executable environments and training agents to coordinate long-horizon work with state maintenance and recovery.
Pengfei Zhou, Hexin Wang, Zhengfeiyang Zhang +5 authors
Game engines provide executable verification and long-horizon trajectories for reinforcement learning post-training of spatial world models, motivating a human-engine verification paradigm.
Zhifei Xie, Jiaqi Lang, Ze An +7 authors
VoiceMem introduces a dual-brain streaming memory architecture for speech language models that improves retrieval accuracy, emotional personalization, and real-time efficiency.
Xuan He, Cong Wei, Yuhao Cheng +20 authors
VGI-bench evaluates visual reasoning in video generation models through 27 tasks, revealing limited reliability and self-correction capabilities.
Liangcai Su, Zhaopeng Feng, Zhuo Chen +13 authors
FrontierChallenge evaluates end-to-end scientific workflows across domains, revealing that frontier models complete only about 20% of tasks despite high partial scores and frequent claims of completion.
Yuandong Pu, Le Zhuo, Sayak Paul +11 authors
The study formalizes probabilistic alignment for world models, introduces PAWBench and PAWEval to evaluate video generators as stochastic samplers, and finds current models fail to match reference behavior distributions.
Zihao Wu, Hongyao Tang, Yi Ma +9 authors
Off-policy reinforcement learning stabilizers vary with data availability, motivating regime-aware algorithms that adapt normalization and Q-function clipping to improve efficiency across CPU and GPU-parallel training.
Tianjie Ju, Zheng Wu, Yueqing Sun +15 authors
UrbanGround evaluates whether multimodal language model agents can sustain reliable navigation and spatial reasoning in a realistic 3D city replica, revealing that local perceptual skills fail to compose into extended goal-directed behavior.
Yunheng Li, Guohong Mu, Hao Li +4 authors
OraRL improves reinforcement learning post-training for video multimodal language models by integrating oracle rollouts with decoupled advantage estimation and sign-balanced pruning, achieving higher sample efficiency and scalability without chain-of-thought generation.
GigaBrain Team, Angen Ye, Axiang Sun +56 authors
GigaBrain-0.7 is a vision-language-action model that improves embodied generalization via a three-system architecture, large-scale heterogeneous pretraining, and joint alignment training.
Aozhe Wang, Zhengxi Lu, Jianze Wang +8 authors
Test-Time Policy Optimization enables label-free test-time training for mathematical reasoning by asymmetrically distilling agreeing rollouts and penalizing disagreeing ones, matching supervised performance.
Shiyi Zhang, Mushui Liu, Yunze Tong +8 authors
Self-OPD eliminates task-specific teachers in flow matching by using self-explored stochastic branches and normalized advantages to optimize the velocity field for multi-objective alignment.
Songchun Zhang, Yaowei Li, Junhao Zhuang +19 authors
EchoWM is an omnimodal world model that generates synchronized high-resolution video, sound, music, and speech while following continuous 6-DoF navigation trajectories across first- and third-person views.
Xingshan Zeng, Zishan Xu, Boju Zhang +11 authors
Agentic data generation is framed as constrained distribution design over factorized experience tuples, emphasizing execution-grounded accuracy, learner-relative complexity, and diversity rather than scale alone.
Junjie Zhou, Ke Mei, Lei Li +3 authors
WeMM-Embedding is a family of universal multimodal embedding models that align text, images, videos, and interleaved inputs in a shared space, achieving state-of-the-art retrieval and recommendation performance across public benchmarks and large-scale WeChat applications.
Wei Zhou, Xiongwei Zhu, Lingdong Kong +14 authors
DiffusionOPSD uses on-policy self-distillation to turn image-level rewards into explicit intermediate targets for diffusion models, improving alignment efficiency and enabling separate analysis of target construction and policy fitting.
Sungho Park, Wonjoong Kim, Rongyuan Tan +10 authors
AutoSaddler automatically improves LLM agent harnesses via offline failure-driven optimization, boosting performance on long-horizon benchmarks.
Yuyuan Feng, Zhishang Xiang, Chaobin Yang +32 authors
Graph Engineering organizes multi-agent LLM systems through dynamic graph structures to coordinate specialized agents and manage complex, evolving tasks.
Yibo Hu, Yu Qian, Mao Gu +6 authors
TLive-Omni is an omni-modal model for live-commerce that unifies image, video, audio, and text via timestamped token grouping, staged supervised training, and reinforcement fine-tuning with verifiable feedback to enable accurate real-time understanding.
TaoLive AIGC LLM Team, Yuhan Sun, Wenhao Lin +7 authors
Harness-Aware Training enables compact models to adapt to evolving digital-avatar harness configurations with low latency and high accuracy.
Seth Karten, Alex L. Zhang, Kevin Thomas +8 authors
Prime Agent is an open-source harness that uses recursive subagents, persistent computation, and agent-to-agent coordination to extend language models' long-horizon capabilities across coding and reasoning tasks.
Long Cui, Xiaoqian Liu, Qi Qin +4 authors
A hierarchical taxonomy and dense supervision strategy improve diffusion-based image editing through fine-grained concepts, large-scale paired data, and granular evaluation.
Yuncheng Guo, Zhanqiu Zhang, Yiwen Guo +1 authors
GameWAM is a unified world-action model for native video-game control that jointly predicts future visuals and executable keyboard-mouse actions using block-causal flow matching, mode-specific distributions, and block-cycle replanning.
Nayeon Kim, Hojin Lee, Yunju Bak +2 authors
A two-step hyperparameter transfer framework predicts optimal learning rates for large Mixture-of-Experts models by scaling across widths and token budgets, enabling efficient pretraining without costly sweeps.
Yi Zhu, Xiongwei Wu, Qiyi Wang +8 authors
MobilePA-Bench is an interactive sandbox benchmark that evaluates mobile planning agents on tool-calling, sub-agent collaboration, memory usage, and composite skill invocation under real runtime constraints.
Xuteng Zhang, Wenhao Zeng, Xiaodong Gu +5 authors
ParaTempo improves parallel reasoning efficiency by using temporal confidence to dynamically prune, retire, and reallocate reasoning branches without synchronization.
Yunze Tong, Mushui Liu, Canyu Zhao +9 authors
InfinityEdit enables continuous, unbounded video editing by using a lightweight adapter with history, temporal causal, and edit cross-attention modules to extend edits into future streaming frames while maintaining stability.
Yiwen Chen, Guosheng Lin, Chi Zhang
Code World Model separates persistent world dynamics from visual rendering by using a language model to generate executable state updates and a video model to render observations from proxy representations.
Yang Xiao, Yusong Sun, Haoyi Wu +7 authors
PILOT enables live self-improvement by allowing a supervisor to steer active workers and distilling execution experience into reusable skills, improving accuracy and efficiency.
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