Coverage shrinkage after SFT in reasoning models correlates with prevalence of decision-point scenarios in data and can be partially mitigated by targeted data synthesis and diversity-aware decoding.
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
-
Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road
Coverage shrinkage after SFT in reasoning models correlates with prevalence of decision-point scenarios in data and can be partially mitigated by targeted data synthesis and diversity-aware decoding.