Fix easy task seed D0↔D1 so dumb greedy targets wrong district; fix signal thresholds 1b5ce08 RohitChandramouli6618 commited on Apr 7
Raise medium seeds so hospital score reflects outbreak severity d795c8e RohitChandramouli6618 commited on Apr 5
Fix: D0 seed above threshold, faster hospital drain, containment as primary metric 1169cdd RohitChandramouli6618 commited on Apr 5
Fix treatment/recovery calibration, hospital breach reward, efficiency grader, recalibrate seeds 4d57df8 RohitChandramouli6618 commited on Apr 5
Swap grader weights: hospital=45% primary constraint, containment=30%, calibrate environment 1f18d1f RohitChandramouli6618 commited on Apr 4
Lower spread rate to 0.09, fix medium seeds, teach focused allocation strategy d3b81d6 RohitChandramouli6618 commited on Apr 4
Raise seeds to prevent premature termination on medium and hard 33a1e6b RohitChandramouli6618 commited on Apr 4
Raise treatment reduction, lower medium/hard seeds, focus prompt strategy 0954bcf RohitChandramouli6618 commited on Apr 4
Fix speed score bug, restore partial resource replenishment, raise easy seed 9cdea7c RohitChandramouli6618 commited on Apr 4
Fix resource replenishment, grader efficiency, seed calibration, prompt strategy ec747a2 RohitChandramouli6618 commited on Apr 4
Fix 1-step termination bug, fix done field in trajectory, recalibrate seeds bf35524 RohitChandramouli6618 commited on Apr 3
Fix core environment solvability: add treatment reduction, calibrate seeds and spread rates, fix hospital grader denominator 455b146 RohitChandramouli6618 commited on Apr 3
Reduce hard max_steps to 15 to prevent WebSocket timeout 14e1e76 RohitChandramouli6618 commited on Apr 2