Hi-TTRL regulates the strength of answer agreement during sampling-stage hints, improving test-time reinforcement learning over standard majority-vote TTRL.
, year = 1980, title =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Hi-TTRL: Regulating Consensus with Hints for Test-Time Reinforcement Learning
Hi-TTRL regulates the strength of answer agreement during sampling-stage hints, improving test-time reinforcement learning over standard majority-vote TTRL.