End-to-end leakage test of the submission pipeline (5 blind-test trials, exempt whole-trial operations frozen) CUT: labels at t < Tc - 1 s that changed when everything from Tc on was replaced by noise (must be 0); earliest change relative to Tc (s; >= -1 is compliant) trials_x_cuts changed_before in_guess in_decided earliest_minus_tc_s combo guess b3v2 causal 15 0.0 0.0 0.0 -0.25 organisers 15 960.0 960.0 0.0 -150.00 eq3 causal 15 0.0 0.0 0.0 -0.75 organisers 15 960.0 960.0 0.0 -150.00 ref_t_bm causal 15 0.0 0.0 0.0 -0.75 organisers 15 960.0 960.0 0.0 -150.00 ref_t_bm_stem20 causal 15 0.0 0.0 0.0 -0.50 organisers 15 960.0 960.0 0.0 -150.00 ref_t_bm_stem50 causal 15 0.0 0.0 0.0 0.00 organisers 15 960.0 960.0 0.0 -150.00 SWAP: share of labels that flip when the two streams are exchanged (decided part; 1 = symmetric) flip_decided flip_guess combo guess b3v2 causal 1.0000 0.0 organisers 1.0000 0.0 eq3 causal 1.0000 0.0 organisers 1.0000 0.0 ref_t_bm causal 1.0000 0.0 organisers 1.0000 0.0 ref_t_bm_stem20 causal 0.7495 0.0 organisers 0.7495 0.0 ref_t_bm_stem50 causal 0.5731 0.0 organisers 0.5731 0.0