VideoWeaver is an agent harness and benchmark for long‑video generation that evaluates and evolves reusable skills. It allows an agent to dynamically compose foundation skills into a workflow based on a single high‑level instruction, rather than following a static pipeline. The benchmark includes 16 task categories and 285 cases with multimodal references, and an evidence‑grounded agent‑as‑judge inspects execution traces and final videos to guide skill evolution, improving both process and output quality.
By Jianhui Wei, Yan Zhang, Jie Tan, Hengchuan Zhu, Xiaotian Zhang, Ziyi Chen, Daoan Zhang, Wei Xu, Yeying Jin, Zuozhu Liu
arXiv:2610.23599v1 Announce Type: new
Abstract: Large language model (LLM) agents tackle complex tasks through reasoning and interaction with external environments. Agent skills provide persistent, t...
By Yewei Zhao, Derong Xu, Wei Chen, Yin Xu, Peilun Zhou, Xiaohan Wang, Guojun Yin, Wei Lin, Mingjun Xiao
SAGE is a framework that automates storyboard creation by learning and evolving directing rules from expert demonstrations. It attributes each narrative group’s decisions to specific rules, refines those rules with localized feedback, and routes only relevant rules to each group during generation. In tests, SAGE matched professional directors on a rubric and reduced authoring time by over 83%.
By Maolin Ran, Xiaoyang Lu, Jiaqi Liu, Jian Wang, Weiwen Liu, Jianghao Lin, Yong Yu, Weinan Zhang
arXiv:2603. 20667v2 Announce Type: replace-cross Abstract: Existing prompt-optimization techniques rely on local signals, causing poor generalization across tasks.
By Balaji Dinesh Gangireddi, Aniketh Garikaparthi, Manasi Patwardhan, Arman Cohan
Learn2Play Bench is a new benchmark that tests how well large language model agents learn from experience in unfamiliar, text‑based games with novel or counterintuitive rules. The benchmark provides reproducible feedback, automatic scoring, and varied game instances to evaluate learning across repeated attempts and transfer to new situations. Findings show that retaining full action records aids learning, human players outperform agents, and the choice of harness significantly impacts performance and inference cost.
By Yibo Li, Jinhang Qiu, Zhi Zheng, Qianyun Guo, Jiaying Wu, Shuo Ji, Bryan Hooi
arXiv:2602. 11790v2 Announce Type: replace Abstract: Although recent end-to-end video generation models demonstrate impressive performance in visually oriented content creation, they remain limited in scenarios that require strict logical rigor and precise knowledge representation, such as instructional and educational media.
By Lingyong Yan, Jiulong Wu, Dong Xie, Weixian Shi, Deguo Xia, Jizhou Huang