DDL2PropBank Agent: Benchmarking Multi-Agent Frameworks’ Developer Experience Through a Novel Relational Schema Mapping Task
Shafiuddin Rehan Ahmed, Wei Wei · 2026
Read DDL2PropBank Agent: Benchmarking Multi-Agent Frameworks’ Developer Experience Through a Novel Relational Schema Mapping Task by Shafiuddin Rehan Ahmed, Wei Wei (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-11198.txt.
Description
Declarative design does not guarantee that AI can use a frame well. Shafiuddin Rehan Ahmed and Sourabh Deshpande offer AI-assistability, a metric joining structural matching with functional correctness. They test it with DDL2PropBank, a task mapping relational schemas to PropBank rolesets, built same way across ten multi-agent frames.
Results overturn intuition: Agno, with one canonical pattern and convention-aligned API, scores highest, while DSPy, the most declarative by design, scores lowest. Convention matching, not declarative style alone, drives assistability, with a moderate link to correctness. All artifacts are released, giving frame builders a principled way to measure how easily AI assistants can generate correct code for their tools.
Catalog record
| CREATOR | Shafiuddin Rehan Ahmed, Wei Wei |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 77 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-11198