Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.06051 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915803414134784 |
|---|---|
| author | Ma, Kexin Li, Bojun Tang, Yuhua Sun, Liting Jin, Ruochun |
| author_facet | Ma, Kexin Li, Bojun Tang, Yuhua Sun, Liting Jin, Ruochun |
| contents | Episodic memory is a central component of human memory, which refers to the ability to recall coherent events grounded in who, when, and where. However, most agent memory systems only emphasize semantic recall and treat experience as structures such as key-value, vector, or graph, which makes them struggle to represent and retrieve coherent events. To address this challenge, we propose a Character-and-Scene based memory architecture(CAST) inspired by dramatic theory. Specifically, CAST constructs 3D scenes (time/place/topic) and organizes them into character profiles that summarize the events of a character to represent episodic memory. Moreover, CAST complements this episodic memory with a graph-based semantic memory, which yields a robust dual memory design. Experiments demonstrate that CAST has averagely improved 8.11% F1 and 10.21% J(LLM-as-a-Judge) than baselines on various datasets, especially on open and time-sensitive conversational questions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_06051 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CAST: Character-and-Scene Episodic Memory for Agents Ma, Kexin Li, Bojun Tang, Yuhua Sun, Liting Jin, Ruochun Computation and Language Episodic memory is a central component of human memory, which refers to the ability to recall coherent events grounded in who, when, and where. However, most agent memory systems only emphasize semantic recall and treat experience as structures such as key-value, vector, or graph, which makes them struggle to represent and retrieve coherent events. To address this challenge, we propose a Character-and-Scene based memory architecture(CAST) inspired by dramatic theory. Specifically, CAST constructs 3D scenes (time/place/topic) and organizes them into character profiles that summarize the events of a character to represent episodic memory. Moreover, CAST complements this episodic memory with a graph-based semantic memory, which yields a robust dual memory design. Experiments demonstrate that CAST has averagely improved 8.11% F1 and 10.21% J(LLM-as-a-Judge) than baselines on various datasets, especially on open and time-sensitive conversational questions. |
| title | CAST: Character-and-Scene Episodic Memory for Agents |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.06051 |