command: /python -I tools/experiments/coser_rouge_check.py /rouge --json /results/rouge_check.json started: 2026-11-01T14:18:19+00:00 ended: 2026-11-01T14:18:22+00:00 exit status: 0 stored rouge_l: [0.07246376532209046, 0.1059001475702016, 0.0732700110099804] rouge 0.2: [0.035698384818706, 0.07225301950008943, 0.03283856857619172] equal 0/3, within 1e-4 0/3 rouge 0.2.1: [0.035698384818706, 0.07225301950008943, 0.03283856857619172] equal 0/3, within 1e-4 0/3 rouge 0.3.0: [0.06994721288011797, 0.12488068998316187, 0.0915646104724211] equal 0/3, within 1e-4 0/3 rouge 0.3.1: [0.06994721288011797, 0.12488068998316187, 0.0915646104724211] equal 0/3, within 1e-4 0/3 rouge 0.3.2: [0.06994721288011797, 0.12488068998316187, 0.0915646104724211] equal 0/3, within 1e-4 0/3 rouge 1.0.0rc0: [0.5758661859996089, 0.8108108071139156, 0.6064690000575411] equal 0/3, within 1e-4 0/3 rouge 1.0.0rc1: [0.5758661859996089, 0.8108108071139156, 0.6064690000575411] equal 0/3, within 1e-4 0/3 rouge 1.0.0rc2: [0.10486891052104824, 0.1737089164028302, 0.14498933607866854] equal 0/3, within 1e-4 0/3 rouge 1.0.0: [0.10486891052104824, 0.1737089164028302, 0.14498933607866854] equal 0/3, within 1e-4 0/3 rouge 1.0.1: [0.10486891052104824, 0.1737089164028302, 0.14498933607866854] equal 0/3, within 1e-4 0/3