RMCT matches the rate of target behaviors like bias-following across input perturbations to reduce sycophancy in LLMs while preserving verbalization of bias cues.
Emrullah and Papailiopoulos, Dimitris and Oymak, Samet , month = feb, year =
1 Pith paper cite this work, alongside 11 external citations. Polarity classification is still indexing.
1
Pith paper citing it
11
external citations · external index
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Consistency Training while Mitigating Obfuscation via Rate Matching
RMCT matches the rate of target behaviors like bias-following across input perturbations to reduce sycophancy in LLMs while preserving verbalization of bias cues.