TaskGen: A Task-Based, Memory-Infused Agentic Framework using StrictJSON
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914881534427136 |
|---|---|
| author | Tan, John Chong Min Saroj, Prince Runwal, Bharat Maheshwari, Hardik Sheng, Brian Lim Yi Cottrill, Richard Chona, Alankrit Kumar, Ambuj Motani, Mehul |
| author_facet | Tan, John Chong Min Saroj, Prince Runwal, Bharat Maheshwari, Hardik Sheng, Brian Lim Yi Cottrill, Richard Chona, Alankrit Kumar, Ambuj Motani, Mehul |
| contents | TaskGen is an open-sourced agentic framework which uses an Agent to solve an arbitrary task by breaking them down into subtasks. Each subtask is mapped to an Equipped Function or another Agent to execute. In order to reduce verbosity (and hence token usage), TaskGen uses StrictJSON that ensures JSON output from the Large Language Model (LLM), along with additional features such as type checking and iterative error correction. Key to the philosophy of TaskGen is the management of information/memory on a need-to-know basis. We empirically evaluate TaskGen on various environments such as 40x40 dynamic maze navigation with changing obstacle locations (100% solve rate), TextWorld escape room solving with dense rewards and detailed goals (96% solve rate), web browsing (69% of actions successful), solving the MATH dataset (71% solve rate over 100 Level-5 problems), Retrieval Augmented Generation on NaturalQuestions dataset (F1 score of 47.03%) |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_15734 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | TaskGen: A Task-Based, Memory-Infused Agentic Framework using StrictJSON Tan, John Chong Min Saroj, Prince Runwal, Bharat Maheshwari, Hardik Sheng, Brian Lim Yi Cottrill, Richard Chona, Alankrit Kumar, Ambuj Motani, Mehul Artificial Intelligence Multiagent Systems TaskGen is an open-sourced agentic framework which uses an Agent to solve an arbitrary task by breaking them down into subtasks. Each subtask is mapped to an Equipped Function or another Agent to execute. In order to reduce verbosity (and hence token usage), TaskGen uses StrictJSON that ensures JSON output from the Large Language Model (LLM), along with additional features such as type checking and iterative error correction. Key to the philosophy of TaskGen is the management of information/memory on a need-to-know basis. We empirically evaluate TaskGen on various environments such as 40x40 dynamic maze navigation with changing obstacle locations (100% solve rate), TextWorld escape room solving with dense rewards and detailed goals (96% solve rate), web browsing (69% of actions successful), solving the MATH dataset (71% solve rate over 100 Level-5 problems), Retrieval Augmented Generation on NaturalQuestions dataset (F1 score of 47.03%) |
| title | TaskGen: A Task-Based, Memory-Infused Agentic Framework using StrictJSON |
| topic | Artificial Intelligence Multiagent Systems |
| url | https://arxiv.org/abs/2407.15734 |