Learning When to Trust in Contextual Social Bandits

Majid Ghasemi [email protected], Mark Crowley [email protected] · 2026

Read Learning When to Trust in Contextual Social Bandits by Majid Ghasemi [email protected], Mark Crowley [email protected] (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-2603-13356.txt.

Description

Majid Ghasemi and Mark Crowley study when to trust feedback in social bandits. Most robust methods assume a source is trustworthy in all contexts, or corrupt only up to a set budget. They name a new failure mode, contextual sycophancy. In it, raters are honest in safe contexts but biased in critical ones. A proof shows that social feedback alone cannot solve this problem. Sparse ground-truth audits are enough, though. Their ESA method learns from audits when to trust each rater.

Catalog record

CREATORMajid Ghasemi [email protected], Mark Crowley [email protected]
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE34553
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.13356

Browse the full library · About the Tower of Babel