{"sources": [
{"url": "https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/tell/README.md", "quote": "Paired: the test adds **+3.21 se 0.02** to the policy the ports implement (`ref/gauss_qm.json`), and is 0.01 short of the true flag."},
{"url": "https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/tell/README.md", "quote": "So are tables solved for a belief that knows its flags, and a cheaper price filter between pins: the test leaves 1.73 to the true price."},
{"url": "https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/solved/real.py", "quote": "sd = sd * self.o.get('sdscale', 1.0)"},
{"url": "https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/tell/tellpol.py", "quote": "On the grid: RealPolicy's own posterior, pi = l_arb / (l_arb + l_ret), with l_ret cut by the chance."},
{"url": "https://api.imd.fun/artifacts/f5a140f51df99423d6ede60b618dbc6eabc37a3050ac439c80dc3824191e9f59", "quote": "stalerival:Policy; tellpol.TellPolicy soft=1 qm=qm_k1e-5.json"}
]}