Procedural Environment Generation for Tool-Use Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908556275482624 |
|---|---|
| author | Sullivan, Michael Hartmann, Mareike Koller, Alexander |
| author_facet | Sullivan, Michael Hartmann, Mareike Koller, Alexander |
| contents | Although the power of LLM tool-use agents has ignited a flurry of recent research in this area, the curation of tool-use training data remains an open problem$-$especially for online RL training. Existing approaches to synthetic tool-use data generation tend to be non-interactive, and/or non-compositional. We introduce RandomWorld, a pipeline for the procedural generation of interactive tools and compositional tool-use data. We show that models tuned via SFT and RL on synthetic RandomWorld data improve on a range of tool-use benchmarks, and set the new SoTA for two metrics on the NESTFUL dataset. Further experiments show that downstream performance scales with the amount of RandomWorld-generated training data, opening up the possibility of further improvement through the use of entirely synthetic data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_11045 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Procedural Environment Generation for Tool-Use Agents Sullivan, Michael Hartmann, Mareike Koller, Alexander Machine Learning I.2.7 Although the power of LLM tool-use agents has ignited a flurry of recent research in this area, the curation of tool-use training data remains an open problem$-$especially for online RL training. Existing approaches to synthetic tool-use data generation tend to be non-interactive, and/or non-compositional. We introduce RandomWorld, a pipeline for the procedural generation of interactive tools and compositional tool-use data. We show that models tuned via SFT and RL on synthetic RandomWorld data improve on a range of tool-use benchmarks, and set the new SoTA for two metrics on the NESTFUL dataset. Further experiments show that downstream performance scales with the amount of RandomWorld-generated training data, opening up the possibility of further improvement through the use of entirely synthetic data. |
| title | Procedural Environment Generation for Tool-Use Agents |
| topic | Machine Learning I.2.7 |
| url | https://arxiv.org/abs/2506.11045 |