Evaluating Model Explanations without Ground Truth
Kaivalya Rawal, Zihao Fu, Eoin Delaney, Chris Russell · 2025
Read Evaluating Model Explanations without Ground Truth by Kaivalya Rawal, Zihao Fu, Eoin Delaney, Chris Russell (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-2505-10399.txt.
Description
Kaivalya Rawal, Chris Russell, and co-authors give a way to grade model tales when no true tale exists to check against, in a 2025 paper. One model call can have many tales that clash, so picking the best is hard. Past grades need a gold tale or just test how touchy the model is.
Their frame needs no gold tale and no touch test, yet still ranks tales on their own worth. It rests on three rules the team sets for fair grades. Tests show it matches past grades and can catch tales bent to fake fairness. Code is shared in public for free.
Catalog record
| CREATOR | Kaivalya Rawal, Zihao Fu, Eoin Delaney, Chris Russell |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-10399