A weakly supervised PRM training method using outcome-only pseudo-labels plus a buffer probability reaches 53.0% F1 on ProcessBench, beating supervised baselines in the paper's comparisons.
Alphamath almost zero: Process super- vision without process
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
FreePRM: Training Process Reward Models Without Ground Truth Process Labels
A weakly supervised PRM training method using outcome-only pseudo-labels plus a buffer probability reaches 53.0% F1 on ProcessBench, beating supervised baselines in the paper's comparisons.