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。
904 篇论文 · 按点赞排序
Hao Liang, Qihan Lin, Meiyi Qiang +5 authors
OmniEdu is a family of instruction-tuned educational foundation models that improve K-12 problem solving, curriculum understanding, and tutoring through capability-balanced training data.
Xiangyu Zeng, Yuandong Yang, Zhiqiu Zhang +21 authors
OneStreamer uses proactive generation to build reusable video memory and respond promptly, achieving top performance on streaming video benchmarks.
Tianming Liang, Junyi Pan, Yuxi Li +5 authors
WeVisOne-VL unifies object-level and pixel-level visual perception via structured token prediction, supporting detection, localization, counting, OCR, segmentation, and matting with a lightweight pixel decoder and multi-stage training.
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.
Pengqi Lu
Deep diffusion transformers face structural instability at extreme depths due to mean-dominated collapse triggered by mean mode screaming, which is mitigated through mean-variance split residuals that maintain stable training while preserving performance.
Jiahao Wang, Hongjin Qian, Yuyang Hu +2 authors
AgentSWE evaluates whether coding agents build software that truly meets user requirements across creation, editing, and optimization stages, revealing widespread failures to deliver faithful behavior.
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.
Ruhan Wang, Yucheng Shi, Zongxia Li +7 authors
The Harness Handbook and Behavior-Guided Progressive Disclosure automatically map agent behaviors to source code, improving localization and edit planning for evolving AI harnesses.
Meijia Chen, Hao Li, Zheng Lu +12 authors
CrossFit mitigates co-cheating in self-evolving search agents by cross-fitting proposal scoring across partitioned sources, substantially reducing false agreement and improving benchmark performance.
DeepSeek-AI, Anyi Xu, B. Li +589 authors
DeepSeek-V4.1-Flash is a multimodal MoE model using a CED architecture and aggressive KV cache compression to lower compute, storage, and bandwidth costs for long-context agent workloads.
Shih-Yang Liu, Xin Dong, Ximing Lu +10 authors
Multi-reward reinforcement learning suffers from reward normalization collapse in GRPO, which GDPO addresses by decoupling reward normalization for improved training stability and performance across reasoning tasks.
Meng Chu, Xuan Billy Zhang, Kevin Qinghong Lin +39 authors
World models are categorized into three capability levels and four law regimes to better understand and develop predictive environment models for AI agents across diverse domains.
Dawei Zhu, Rui Meng, Yale Song +4 authors
_paperbanana is an agentic framework that automates the creation of publication-ready academic illustrations using advanced vision-language models and image generation techniques.
Fuyao Zhang, Wenwen He, Jiaming Zhang +6 authors
FocusGUI is a lightweight adapter that dynamically selects relevant multimodal evidence for GUI agents to prevent compounding errors from irrelevant context, improving action accuracy without fine-tuning the backbone.
Fang Wu, Da Xing, Yanjie Huang +16 authors
UniEvo-VL enables multimodal models to self-improve during inference by using internal critiques to guide diffusion-based generation without external teachers.
Xi Xiao, Tianchen Zhao, Youngeun Kim +10 authors
ReaLVR improves latent visual reasoning by supervising latent tokens with visual evidence and answer contrast, scaling robustly to 235B parameters.
Xuying Ning, Katherine Tieu, Dongqi Fu +39 authors
Large language models are increasingly used as operational substrates for agent reasoning and execution in agentic systems, with code serving as a unified infrastructure layer across multiple domains and applications.
Mind Lab, Song Cao, Vic Cao +59 authors
MinT is a managed infrastructure system that enables efficient low-rank adaptation training and serving by keeping base models resident and moving lightweight adapter revisions, scaling across multiple dimensions including large model architectures, reduced storage requirements, and distributed policy management.
Bingxi Hou, Guochao Jiang, Guofeng Quan +4 authors
Waypoint-OPD improves policy-shift distillation by gating endpoint shifts using directional agreement between early and late teacher policy changes, boosting reasoning performance with minimal overhead.
Zihao Li, Jiaru Zou, Feihao Fang +6 authors
Eywa is a heterogeneous agentic framework that extends language-centric systems to scientific foundation models by integrating domain-specific models with language-based reasoning interfaces for improved performance across diverse scientific domains.
Guobin Shen, Chenxiao Zhao, Xiang Cheng +2 authors
VESPO addresses training instability in LLM reinforcement learning by using variational formulation with variance reduction to correct policy divergence without length normalization.
Koutian Wu, Junjie Zhou, Ergan Shang +4 authors
Benchmark Radar is a searchable living database and discovery engine for AI evaluation benchmarks that aggregates sources, score histories, and evidence to support benchmark selection and comparison.
Peng Xia, Rujun Han, Zifeng Wang +11 authors
Regularized recursive self-improvement of agent harnesses improves out-of-distribution performance by constraining edit proposals and selections to favor reusable mechanisms over benchmark-specific memorization.
Haoyu Zhao, Zihao Zhang, Xudong Wang +4 authors
MaLiang-Harness defines the Program-to-Visual gap and uses persistent executable generation, traceable editing, and revision-aware verification to improve multimodal language model-driven visual creation.
Chuyang Wei, Haoxiang Guan, Minghang Zhu +5 authors
A framework for self-evolving agent harnesses uses feedback-driven revisions and evolutionary optimization with theoretical generalization guarantees while keeping model weights fixed.
Guanting Dong, Xinyu Yang, Jiajie Jin +9 authors
Agent-Reflex trains interactive agents to reflect and recover from noisy tool feedback and user corrections through skill discovery, supervised fine-tuning, and on-policy optimization.
Dingyu Yao, Junhao Zhou, Chenxu Yang +12 authors
A vision-language model operates continuously in real-time, making autonomous decisions about when to respond or delegate, enabling interactive systems that perceive and act upon environmental changes without user prompting.
Chengwen Liu, Xiaomin Yu, Zhuoyue Chang +13 authors
VideoDR benchmark enables video question answering by combining cross-frame visual extraction, web retrieval, and multi-hop reasoning in open-domain settings.
Hao Liang, Mingrui Chen, Hengyi Feng +2 authors
DataFlex-RL evaluates reinforcement learning data policies and finds that uniform sampling matches or exceeds adaptive rollout selection, reweighting, and domain mixing across math, logic, and science benchmarks.
Bingxi Hou, Guochao Jiang, Guofeng Quan +4 authors
On-policy distillation with strict token alignment outperforms broader cross-tokenizer supervision by avoiding weak or conflicting gradient signals.
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