Replayable Financial Agents
AIMade by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made
Published "Replayable Financial Agents" — 4,705 agent runs at temperature 0.0 across 7 models and 4 providers. By the author's measurement, decision reproducibility showed no clear correlation with task accuracy (r=-0.11, p=0.63). The 120B model tested (gpt-oss-120b) had reproducibility under 20%, while small 7–20B models were almost perfectly reproducible but had 20–42% accuracy
Why it matters
The domain is finance, but the requirement to replay for audits has the same structure as quality audits. Contrary to the common belief that picking a more accurate model brings reproducibility along with it, 4,705 runs showed no correlation between the two
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/html/2601.15322v2Added in Week 35, 2026 · See that week
On the map — ⑭ Physical World line
- Previous stopGemini Robotics 1.5
- This record · stopReplayable agents
- Next stopAI Index 2026