Daily AI Papers — August 28, 2026

1. Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World Models

Authors: Pengfei Zhou, Hexin Wang, Zhengfeiyang Zhang, Yixing Ma, Zhenglin Wan, Kaipeng Zhang, Wangbo Zhao, Yang You arXiv: arxiv.org/abs/2608.25518 Summary: A common strategy for scaling world models is to train on more crawled video with more compute. We argue that this strategy is inefficient: scaling world models also requires a recursive data engine that offers grounded reward signals. Trending because: 113 HuggingFace upvotes + strong community engagement on the topic

2. PAWBench: How Far Are We from Probabilistically Aligned World Modeling?

Authors: Yuandong Pu, Le Zhuo, Sayak Paul, Gabriel Jorge Menezes, Avram Đorđević, Shiyang Li, Yifan Zhou, Bin Fu, Wenlong Zhang, Junjun He, Yu Qiao, Yihao Liu, Jingbo Xing, Xi Chen arXiv: arxiv.org/abs/2608.27345 Summary: Recent video generation models are increasingly framed as world models. Many physical processes can unfold in more than one valid way. Trending because: 73 HuggingFace upvotes + strong community engagement on the topic

3. UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City

Authors: Tianjie Ju, Zheng Wu, Yueqing Sun, Yuhan Cui, Bobo Li, Shengqiong Wu, Pengzhou Cheng, Haodong Zhao, Zongru Wu, Xinbei Ma, Doris Zhang, Kunling Li, Mong-Li Lee, Wynne Hsu, Hao Fei, Qi Gu, Gongshen Liu, Zhuosheng Zhang arXiv: arxiv.org/abs/2608.27456 Summary: Multimodal large language models (MLLMs) can interpret a street view, but urban agency depends on whether such local evidence remains useful after the agent starts to move. In this paper, we investigate how far current MLLM agents can turn local urban perception into reliable action in a complicated real-scale city. Trending because: 66 HuggingFace upvotes + strong community engagement on the topic

4. TTPO: Test-Time Policy Optimization

Authors: Aozhe Wang, Zhengxi Lu, Jianze Wang, Shangke Lv, Ying Liu, Weiming Lu, Jun Xiao, Yueting Zhuang, Hua Yang, Qianglong Chen, Yongliang Shen arXiv: arxiv.org/abs/2608.27448 Summary: Recent prominent post-training methods, such as Reinforcement Learning (RL) and On-Policy Self-Distillation (OPSD), have driven rapid progress in mathematical reasoning for large language models, yet their reliance on ground-truth labels precludes test-time training (TTT). Replacing ground truth with majority-vote pseudo-labels is a natural alternative, yet it is fragile: an incorrect vote corrupts the teacher and misleads every token. Trending because: 58 HuggingFace upvotes + strong community engagement on the topic

5. What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents

Authors: Xingshan Zeng, Zishan Xu, Boju Zhang, Yuzhou Wu, Lingzhi Wang, Jianghao Lin, Liangyou Li, Yasheng Wang, Lifeng Shang, Xin Jiang, Weinan Zhang, Yong Yu, Qun Liu, Weiwen Liu arXiv: arxiv.org/abs/2608.27260 Summary: LLM agents increasingly rely on generated interaction data to learn how to interact with external environments. Agentic data generation must maintain consistency among environments, tasks, interactions, and success signals while producing experience that is useful rather than merely abundant. Trending because: 55 HuggingFace upvotes + strong community engagement on the topic

6. Self-OPD: On-Policy Distillation for Flow Matching Models without Teacher

Authors: Shiyi Zhang, Mushui Liu, Yunze Tong, Wanggui He, Siyu Zou, Jinlong Liu, Yunlong Yu, Jian Song, Hao Jiang, Pipei Huang, Bo Zheng arXiv: arxiv.org/abs/2608.26872 Summary: On-policy distillation (OPD), which leverages a pre-trained, specialized teacher model to provide dense supervisory signals, has achieved significant success in Large Language Models (LLMs) and has recently been adapted to flow matching models. However, this paradigm suffers from two major issues: First, training a separate, task-specific teacher for every new objective incurs high computational costs. Trending because: 55 HuggingFace upvotes + strong community engagement on the topic

7. Training Agents to Evolve with Their Harness: TaoLive Digital Avatar Agent Technical Report

Authors: TaoLive AIGC LLM Team, Yuhan Sun, Wenhao Lin, Yongdong Luo, Yibo Hu, Meiguang Jin, Junfeng Ma, Weihang Pan, Jiaxin Zhao, Zulong Chen arXiv: arxiv.org/abs/2608.15763 Summary: AI-powered digital avatar streamers must answer product questions, engage viewers, and execute marketing strategies in real time, demanding low latency, frequent strategy updates, and accurate yet effective responses. Evolvable Harnesses, whose Skills, Hooks, prompts, and tools can be updated independently of model weights, enable rapid iteration but expose a trade-off: large models adapt zero-shot yet are too slow, whereas compact models meet latency targets but overfit to fixed Harness configurations. Trending because: 42 HuggingFace upvotes + strong community engagement on the topic

8. GameWAM: A World Action Model for Video Games

Authors: Yuncheng Guo, Zhanqiu Zhang, Yiwen Guo, Weijia Li arXiv: arxiv.org/abs/2608.26200 Summary: Modern video games combine first-person perception, rapid visual changes, persistent world state, and heterogeneous native controls. Existing game agents map visual and task context directly to actions but lack explicit world dynamics modeling, whereas interactive game world models predict visual futures from supplied actions but do not serve as task policies. Trending because: 36 HuggingFace upvotes + strong community engagement on the topic

9. PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents

Authors: Yang Xiao, Yusong Sun, Haoyi Wu, Wenyang Hui, Wen Da, Zhaokai Luo, Mu Chuan, Yao Hu, Wenjie Li, Chengyue Jiang arXiv: arxiv.org/abs/2608.26530 Summary: Long-horizon agent runs generate experience that can improve both the current run and future work. Most self-improvement methods process this experience only after execution ends, so they cannot redirect the active run or immediately apply and validate lessons learned from it. Trending because: 25 HuggingFace upvotes + strong community engagement on the topic

10. Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization

Authors: Jiaming Zhou, Qihang Zhang, Gangwei Xu, Cunxin Fan, Yujie Zhao, Ruilin Wang, Yiming Luo, Shuai Yang, Xing Zhu, Yujun Shen, Junwei Liang, Yinghao Xu arXiv: arxiv.org/abs/2608.26103 Summary: Zero-shot cross-task generalization, where a policy must execute manipulation tasks never seen during training, remains a central challenge in robot learning. In large language models, a novel task can be performed simply by specifying it in the context, without any parameter update. Trending because: 16 HuggingFace upvotes + strong community engagement on the topic

11. Understanding Evolution Strategies for LLM Reasoning: Broader Reasoning Coverage than GRPO

Authors: Yunpeng Ba, Zhi Zheng, Yue Xie, Jiaqing Li, Xialiang Tong, Tao Zhong, Mingxuan Yuan, Zhichao Lu, Xuyang Wu, Zhenkun Wang arXiv: arxiv.org/abs/2608.27351 Summary: Evolution Strategies (ES) have recently emerged as a memory-efficient post-training paradigm for LLM reasoning. However, the optimization behavior of ES remains understudied, making it hard to define its advantage scope compared to mainstream post-training paradigms (e.g., Group Relative Policy Optimization (GRPO)). Trending because: 13 HuggingFace upvotes + strong community engagement on the topic

12. A Programming Paradigm for Spatiotemporal Composability

Authors: Yifan Shi, Wei Zhang, Tianyi Cui arXiv: arxiv.org/abs/2608.25512 Summary: Modern software – from plugin systems to self-evolving agent harnesses – increasingly requires dynamic composition, yet its formal foundations remain underdeveloped. We identify two orthogonal dimensions of the problem: temporal composability, the ability to completely revert a component’s side effects upon removal, and spatial composability, the ability to declare and reactively manage inter-component dependencies. Trending because: 11 HuggingFace upvotes + strong community engagement on the topic

13. MA-VLA: Multi-Arm Vision-Language-Action Model for Collaboration and Compositional Generalization

Authors: Zaibin Zhang, Junlan Xiao, Zhongbo Zhang, Yifan Wang, Li Kang, Yiran Qin, Changxing Xia, Heng Zhou, Talas Fu, Enshen Zhou, Ruimao Zhang, Zhenfei Yin, Huchuan Lu, Lijun Wang arXiv: arxiv.org/abs/2608.25864 Summary: Multi-arm collaboration is becoming a core capability in embodied manipulation. Recent vision-language-action (VLA) models integrate perception, language, and control, but most represent language as a single global instruction and do not provide an explicit mechanism for assigning and composing arm-specific behaviors. Trending because: 8 HuggingFace upvotes + strong community engagement on the topic

14. GUI-Primitives: Diagnosing Spatial Reasoning Failures in Vision-Language GUI Grounding

Authors: Md Abrar Jahin, Md Rizwan Parvez arXiv: arxiv.org/abs/2608.21832 Summary: Computer-use agents ground natural-language instructions in screenshots to locate interface elements, yet existing benchmarks do not isolate whether models bind relational language to the correct element. We introduce GUI-Primitives, a 994-item benchmark of contrastive instruction pairs over seven spatial relations in graphical user interfaces (left/right, above/below, containment, alignment, proximity, list ordinal, occlusion). Trending because: 8 HuggingFace upvotes + strong community engagement on the topic

15. Gated Recurrent Transformers: Expressive Depth through Recurrent Modulation

Authors: Amr Hegazy, Amr Alanwar, Mostafa Elhoushi arXiv: arxiv.org/abs/2608.15062 Summary: Scaling transformer language models creates an inherent tension between expressivity and memory efficiency. While unique weights across layers preserve functional specialization—from input-grounding to abstract refinement—they incur a substantial memory footprint. Trending because: 8 HuggingFace upvotes + strong community engagement on the topic

16. WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution

Authors: Liyan Tang, Cyrus Rashtchian, Chun-Sung Ferng, Andrew Tomkins, Da-Cheng Juan, Tu Vu arXiv: arxiv.org/abs/2608.27454 Summary: Agent skills package specialized knowledge and workflows into reusable resources that extend AI agent capabilities. Recent work automatically discovers such skills from agent experience, which enables agents to progressively adapt through interaction. Trending because: 7 HuggingFace upvotes + strong community engagement on the topic

17. Magpie: Real-Time World Renderer for Interactive Games

Authors: Xiaoyu Zhan, Xinyu Wang, Xiaohong Zhang, Huanjie Zhu, Tengjiao Sun, Pengcheng Fang, Jiaxing Yu, Yanwen Guo, Dongjie Fu arXiv: arxiv.org/abs/2608.27168 Summary: Modern game development relies heavily on conventional graphics pipelines. High-quality visual content requires modeling, material authoring, animation, lighting, effects, and runtime optimization, making asset production expensive and extending the development cycle of game prototypes. Trending because: 6 HuggingFace upvotes + strong community engagement on the topic

18. CaRGo-T: Causal Reasoning Graph-of-Thought improves Multimodal Humor Comprehension

Authors: Abhilash Nandy, Rahul Seetharaman, Aman Bansal, Rounak Saha, Manav Nitin Kapadnis, Millon Madhur Das, Pawan Goyal, Niloy Ganguly arXiv: arxiv.org/abs/2608.23172 Summary: Large-scale vision-language models (VLMs) have demonstrated remarkable versatility across a wide range of multimodal tasks. However, understanding humor remains challenging because humorous content often depends on subtle interactions among entities, events, context, and implicit relationships across image and text modalities. Trending because: 5 HuggingFace upvotes + strong community engagement on the topic

19. Procedura: Agentic 3D Modeling with Procedural Control

Authors: Youtian Lin, Yikang Yang, Zhanpeng Hu, Mengqi Zhou, Feihu Zhang, Xun Cao, Jiaheng Liu, Yao Yao arXiv: arxiv.org/abs/2608.26238 Summary: Native 3D generators now recover impressive mesh geometry from a single image. However, a dense mesh stays soft where a machined object should be sharp, it carries no part decomposition, and it exposes no parameter a user could edit. Trending because: 5 HuggingFace upvotes + strong community engagement on the topic

20. Pushing the Limits of High-Resolution Weather Forecasting through Data Scaling

Authors: Yang Zhao, Peisong Niu, Tian Zhou, Ziqing Ma, Guanlong Ma, Rong Jin, Huiling Yuan, Liang Sun arXiv: arxiv.org/abs/2608.14652 Summary: The development of 0.1^{circ} global weather forecasting models based on machine learning (ML) is constrained by the limited availability of high-resolution data, as decades of reanalysis are only available at 0.25^{circ} resolution. While existing approaches fine-tune 0.25^{circ} forecast models on limited 0.1^{circ} samples, we show that this transfer is hindered by the irreversible information loss inherent in coarse-resolution forecasting. Trending because: 4 HuggingFace upvotes + strong community engagement on the topic