n_matched_items: 300
n_seeds: 1
base_accuracy: 0.620000
rl_accuracy: 0.650000
accuracy_delta_rl_minus_base: 0.030000
accuracy_delta_ci_lo: -0.013333
accuracy_delta_ci_hi: 0.073333
accuracy_moved_ge_0p05: 0
primary_did: -0.016667
primary_did_ci_lo: -0.053333
primary_did_ci_hi: 0.020000
primary_did_p: 0.488951
primary_did_seed_unit_mean: -0.016667
primary_did_seed_unit_ci_lo: -0.016667
primary_did_seed_unit_ci_hi: -0.016667
primary_did_seed_unit_p: nan
paraphrase_flip_delta_rl_minus_base: -0.076667
paraphrase_flip_delta_ci_lo: -0.116667
paraphrase_flip_delta_ci_hi: -0.036667
paraphrase_flip_delta_p: 0.000400
base_paraphrase_jaccard_mean: 0.646474
rl_paraphrase_jaccard_mean: 0.653536
fidelity_token_overlap_base_minus_rl: -0.007062
fidelity_token_overlap_diff_ci_lo: -0.028987
fidelity_token_overlap_diff_ci_hi: 0.015410
base_paraphrase_rouge_l_mean: 0.755365
rl_paraphrase_rouge_l_mean: 0.768987
fidelity_rouge_l_base_minus_rl: -0.013622
fidelity_rouge_l_diff_ci_lo: -0.036579
fidelity_rouge_l_diff_ci_hi: 0.008839
base_paraphrase_semantic_cosine_mean: 0.952919
rl_paraphrase_semantic_cosine_mean: 0.956337
fidelity_semantic_base_minus_rl: -0.003418
fidelity_semantic_diff_ci_lo: -0.013211
fidelity_semantic_diff_ci_hi: 0.005477
fidelity_semantic_equiv_bound: 0.050000
fidelity_semantic_point_within_bound: 1
fidelity_semantic_ci_within_bound: 1
base_accuracy_ci_lo: 0.563333
base_accuracy_ci_hi: 0.676667
base_intact_flip_rate: 0.130000
base_intact_flip_rate_ci_lo: 0.096667
base_intact_flip_rate_ci_hi: 0.170000
base_paraphrase_flip_rate: 0.143333
base_paraphrase_flip_rate_ci_lo: 0.106667
base_paraphrase_flip_rate_ci_hi: 0.183333
base_scramble_flip_rate: 0.683333
base_scramble_flip_rate_ci_lo: 0.630000
base_scramble_flip_rate_ci_hi: 0.736667
rl_accuracy_ci_lo: 0.593333
rl_accuracy_ci_hi: 0.703333
rl_intact_flip_rate: 0.070000
rl_intact_flip_rate_ci_lo: 0.043333
rl_intact_flip_rate_ci_hi: 0.100000
rl_paraphrase_flip_rate: 0.066667
rl_paraphrase_flip_rate_ci_lo: 0.040000
rl_paraphrase_flip_rate_ci_hi: 0.096667
rl_scramble_flip_rate: 0.710000
rl_scramble_flip_rate_ci_lo: 0.659917
rl_scramble_flip_rate_ci_hi: 0.760000
primary_did_seed_42: -0.016667