SFT on weak demonstrations followed by RL elicits full performance from sandbagging LLMs, but only when training and deployment are indistinguishable to the model.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Removing Sandbagging in LLMs by Training with Weak Supervision
SFT on weak demonstrations followed by RL elicits full performance from sandbagging LLMs, but only when training and deployment are indistinguishable to the model.