← Back
Finding 4360Emerging EvidenceValidation V0

FinZero turns time series into images and trains a 3B-parameter vision-language model with uncertainty-aware RL (UARPO) to reason, predict rises/falls, and assess confidence using Qwen2.5-VL-3B; inputs avoid normalization, window constraints.

75%Confidence
1Evidence objects
v1Version
DraftStatus

Evidence trail

Knowledge status

This Finding was extracted from the configured corpus. It is versioned, traceable, and may evolve through editorial review or new corpus evidence.