Cycle-Consistent Search uses question reconstruction from search trajectories as a reward for training agents, matching supervised performance without gold labels.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Cycle-Consistent Search: Question Reconstructability as a Proxy Reward for Search Agent Training
Cycle-Consistent Search uses question reconstruction from search trajectories as a reward for training agents, matching supervised performance without gold labels.