GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization (V1.0)
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910143734611968 |
|---|---|
| author | Liang, Jiaqing Han, Jinyi Li, Weijia Wang, Xinyi Zhang, Zhoujia Jiang, Zishang Liao, Ying Li, Tingyun Huang, Ying Shen, Hao Wu, Hanyu Guo, Fang Wang, Keyi Hong, Zhonghua Lu, Zhiyu Ma, Lipeng Jiang, Sihang Xiao, Yanghua |
| author_facet | Liang, Jiaqing Han, Jinyi Li, Weijia Wang, Xinyi Zhang, Zhoujia Jiang, Zishang Liao, Ying Li, Tingyun Huang, Ying Shen, Hao Wu, Hanyu Guo, Fang Wang, Keyi Hong, Zhonghua Lu, Zhiyu Ma, Lipeng Jiang, Sihang Xiao, Yanghua |
| contents | Long-horizon large language model (LLM) agents are fundamentally limited by context. As interactions become longer, tool descriptions, retrieved memories, and raw environmental feedback accumulate and push out the information needed for decision-making. At the same time, useful experience gained from tasks is often lost across episodes. We argue that long-horizon performance is determined not by context length, but by how much decision-relevant information is maintained within a finite context budget. We present GenericAgent (GA), a general-purpose, self-evolving LLM agent system built around a single principle: context information density maximization. GA implements this through four closely connected components: a minimal atomic tool set that keeps the interface simple, a hierarchical on-demand memory that only shows a small high-level view by default, a self-evolution mechanism that turns verified past trajectories into reusable SOPs and executable code, and a context truncation and compression layer that maintains information density during long executions. Across task completion, tool use efficiency, memory effectiveness, self-evolution, and web browsing, GA consistently outperforms leading agent systems while using significantly fewer tokens and interactions, and it continues to evolve over time. Project: https://github.com/lsdefine/GenericAgent |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_17091 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization (V1.0) Liang, Jiaqing Han, Jinyi Li, Weijia Wang, Xinyi Zhang, Zhoujia Jiang, Zishang Liao, Ying Li, Tingyun Huang, Ying Shen, Hao Wu, Hanyu Guo, Fang Wang, Keyi Hong, Zhonghua Lu, Zhiyu Ma, Lipeng Jiang, Sihang Xiao, Yanghua Computation and Language Long-horizon large language model (LLM) agents are fundamentally limited by context. As interactions become longer, tool descriptions, retrieved memories, and raw environmental feedback accumulate and push out the information needed for decision-making. At the same time, useful experience gained from tasks is often lost across episodes. We argue that long-horizon performance is determined not by context length, but by how much decision-relevant information is maintained within a finite context budget. We present GenericAgent (GA), a general-purpose, self-evolving LLM agent system built around a single principle: context information density maximization. GA implements this through four closely connected components: a minimal atomic tool set that keeps the interface simple, a hierarchical on-demand memory that only shows a small high-level view by default, a self-evolution mechanism that turns verified past trajectories into reusable SOPs and executable code, and a context truncation and compression layer that maintains information density during long executions. Across task completion, tool use efficiency, memory effectiveness, self-evolution, and web browsing, GA consistently outperforms leading agent systems while using significantly fewer tokens and interactions, and it continues to evolve over time. Project: https://github.com/lsdefine/GenericAgent |
| title | GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization (V1.0) |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2604.17091 |