ML-Based Automata Simplification for Symbolic Accelerators
Tiffany Yu, Rye Stahle-Smith, Darssan Eswaramoorthi, Rasha Karakchi · 2025
Read ML-Based Automata Simplification for Symbolic Accelerators by Tiffany Yu, Rye Stahle-Smith, Darssan Eswaramoorthi, Rasha Karakchi (2025) 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-2507-08751.txt.
Description
Tiffany Yu and co-authors present in 2025 a learning tool that slimmed the pattern graphs run on symbolic accelerator chips. Such chips chew through pattern tasks in genomics, language, and security, but big graphs eat memory and tangle routing. Their tool, AutoSlim, scores each link in the graph and prunes the weak ones with a forest classifier.
Correctness of the pattern task is kept while graph density falls. On test graphs up to sixty-four thousand nodes, chip resource use fell by up to forty percent and links by over thirty percent. The tool also scaled to graphs ten times larger than past test sets, a step toward bigger jobs.
Catalog record
| CREATOR | Tiffany Yu, Rye Stahle-Smith, Darssan Eswaramoorthi, Rasha Karakchi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 17 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-08751