Complexity Horizons of Compressed Models in Analog Circuit Analysis
Pacome Simon Mbonimpa · 2026
Read Complexity Horizons of Compressed Models in Analog Circuit Analysis by Pacome Simon Mbonimpa (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-2605-02285.txt.
Description
Pacome Simon Mbonimpa proposes choosing the smallest compressed language model that can still handle a circuit analysis task, based on where its knowledge stops. Deploying large models for engineering faces a trade-off between reasoning accuracy and compute cost. Flat performance scores hide the layered structure of electronics knowledge, where later concepts rest on earlier ones.
The framework arranges circuit concepts as a directed acyclic graph of prerequisites. An agentic pipeline builds prerequisite-based data sets, and an evaluation engine cascades each query across compressed model variants by tier. That reveals each tier's complexity horizon, the level where it begins to fail. The smallest model whose horizon covers a query can then be selected, saving compute without needless loss.
Catalog record
| CREATOR | Pacome Simon Mbonimpa |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.02285