Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.07568 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917071558803456 |
|---|---|
| author | Hsiao, Vincent Roberts, Mark Smith, Leslie |
| author_facet | Hsiao, Vincent Roberts, Mark Smith, Leslie |
| contents | Large language models (LLMs) often struggle when performing agentic tasks without substantial tool support, prom-pt engineering, or fine tuning. Despite research showing that domain-dependent, procedural knowledge can dramatically increase planning efficiency, little work evaluates its potential for improving LLM performance on agentic tasks that may require implicit planning. We formalize, implement, and evaluate an agentic LLM workflow that leverages procedural knowledge in the form of a hierarchical task network (HTN). Empirical results of our implementation show that hand-coded HTNs can dramatically improve LLM performance on agentic tasks, and using HTNs can boost a 20b or 70b parameter LLM to outperform a much larger 120b parameter LLM baseline. Furthermore, LLM-created HTNs improve overall performance, though less so. The results suggest that leveraging expertise--from humans, documents, or LLMs--to curate procedural knowledge will become another important tool for improving LLM workflows. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07568 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Procedural Knowledge Improves Agentic LLM Workflows Hsiao, Vincent Roberts, Mark Smith, Leslie Artificial Intelligence Large language models (LLMs) often struggle when performing agentic tasks without substantial tool support, prom-pt engineering, or fine tuning. Despite research showing that domain-dependent, procedural knowledge can dramatically increase planning efficiency, little work evaluates its potential for improving LLM performance on agentic tasks that may require implicit planning. We formalize, implement, and evaluate an agentic LLM workflow that leverages procedural knowledge in the form of a hierarchical task network (HTN). Empirical results of our implementation show that hand-coded HTNs can dramatically improve LLM performance on agentic tasks, and using HTNs can boost a 20b or 70b parameter LLM to outperform a much larger 120b parameter LLM baseline. Furthermore, LLM-created HTNs improve overall performance, though less so. The results suggest that leveraging expertise--from humans, documents, or LLMs--to curate procedural knowledge will become another important tool for improving LLM workflows. |
| title | Procedural Knowledge Improves Agentic LLM Workflows |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2511.07568 |