architecture · first seen 7 Oct 2026
self-evolving agent harness
Also: SERA
An agentic wrapper that iteratively improves its code and skills around a fixed model using experience.
Mentions per week
PapersCodeDiscussion
Earliest sources
- EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-EvolutionPapers · 7 Oct 2026 · Python Song, Zhixuan Liang, Kelsey Fu, Mengdi Wang, Junfeng Yang, Shilong Liu
- VideoEvoHarness: Self Evolving Agent Harness from Unlabeled Videos for Video UnderstandingPapers · 11 Oct 2026 · Weixin Xu, Yifan Liu, Yi Wang, Mingfei Rong, Zhenyu Yang, Zhisheng Zhong, Shengs…
- From VLM Planners to Robotic Agents: Self-Evolving Harnesses for History-Dependent ManipulationPapers · 21 Oct 2026 · Tingwen Du, Haoyu Guo, Hanxue Zhang, Xuanxu Wang, Junting Dong, Chao Gao, Mingfe…
- EvoGroup: A Self-Evolving Harness for Group Agents in Shared ContextsPapers · 2 Nov 2026 · Xinyi Xu, Xiaoyang Li, Yucheng Feng, Wenqing Chen, Youliang Yuan, Wei Shi, Rongx…
Latest sources
- EvoGroup: A Self-Evolving Harness for Group Agents in Shared ContextsPapers · 2 Nov 2026 · Xinyi Xu, Xiaoyang Li, Yucheng Feng, Wenqing Chen, Youliang Yuan, Wei Shi, Rongx…
- From VLM Planners to Robotic Agents: Self-Evolving Harnesses for History-Dependent ManipulationPapers · 21 Oct 2026 · Tingwen Du, Haoyu Guo, Hanxue Zhang, Xuanxu Wang, Junting Dong, Chao Gao, Mingfe…
- VideoEvoHarness: Self Evolving Agent Harness from Unlabeled Videos for Video UnderstandingPapers · 11 Oct 2026 · Weixin Xu, Yifan Liu, Yi Wang, Mingfei Rong, Zhenyu Yang, Zhisheng Zhong, Shengs…
- EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-EvolutionPapers · 7 Oct 2026 · Python Song, Zhixuan Liang, Kelsey Fu, Mengdi Wang, Junfeng Yang, Shilong Liu