{ "STSBenchmark": { "train": { "pearson": [ 0.7675016313639117, 0.0 ], "spearman": [ 0.7404488937353225, 0.0 ], "nsamples": 5749 }, "dev": { "pearson": [ 0.7658521041744508, 1.4020329449305176e-289 ], "spearman": [ 0.7730453513050343, 2.066561283121993e-298 ], "nsamples": 1500 }, "test": { "pearson": [ 0.7623056519318484, 1.8448021245198884e-262 ], "spearman": [ 0.7572290027800557, 5.3120116722114526e-257 ], "nsamples": 1379 }, "all": { "pearson": { "all": 0.7647570029076096, "mean": 0.7652197958234037, "wmean": 0.7663843913985655 }, "spearman": { "all": 0.7534738635333297, "mean": 0.7569077492734708, "wmean": 0.7487978108339844 } } }, "SICKRelatedness": { "train": { "pearson": [ 0.7707517065205873, 0.0 ], "spearman": [ 0.6949522587450165, 0.0 ], "nsamples": 4500 }, "dev": { "pearson": [ 0.778349180191507, 9.711324138536703e-103 ], "spearman": [ 0.7176593438481393, 2.743332620783351e-80 ], "nsamples": 500 }, "test": { "pearson": [ 0.7642987657770804, 0.0 ], "spearman": [ 0.6847455515560564, 0.0 ], "nsamples": 4927 }, "all": { "pearson": { "all": 0.7678373187987577, "mean": 0.7711332174963915, "wmean": 0.7679316297393041 }, "spearman": { "all": 0.6909477744903625, "mean": 0.6991190513830707, "wmean": 0.6910301368785469 } } }, "MR": { "devacc": 81.19, "acc": 80.75, "ndev": 10662, "ntest": 10662 }, "CR": { "devacc": 87.12, "acc": 86.73, "ndev": 3775, "ntest": 3775 }, "SUBJ": { "devacc": 95.61, "acc": 95.29, "ndev": 10000, "ntest": 10000 }, "MPQA": { "devacc": 89.04, "acc": 88.61, "ndev": 10606, "ntest": 10606 }, "SST2": { "devacc": 85.21, "acc": 85.61, "ndev": 872, "ntest": 1821 }, "TREC": { "devacc": 80.69, "acc": 86.8, "ndev": 5452, "ntest": 500 }, "MRPC": { "devacc": 71.69, "acc": 70.32, "f1": 79.57, "ndev": 4076, "ntest": 1725 }, "STS12": { "MSRpar": { "pearson": [ 0.4514113169306324, 6.251817668045887e-39 ], "spearman": [ 0.4499549654577979, 1.1613312893486279e-38 ], "nsamples": 750 }, "MSRvid": { "pearson": [ 0.8599559900295872, 1.0643859926359198e-220 ], "spearman": [ 0.8649974875962446, 3.229629761923383e-226 ], "nsamples": 750 }, "SMTeuroparl": { "pearson": [ 0.4931333562119396, 1.6812261260529903e-29 ], "spearman": [ 0.6128103790153274, 1.1381831791276455e-48 ], "nsamples": 459 }, "surprise.OnWN": { "pearson": [ 0.7365214189654901, 3.947242882144461e-129 ], "spearman": [ 0.683719802480883, 1.910512439440496e-104 ], "nsamples": 750 }, "surprise.SMTnews": { "pearson": [ 0.6084967943558258, 8.882847362372895e-42 ], "spearman": [ 0.6324009475046439, 5.9238371155257115e-46 ], "nsamples": 399 }, "all": { "pearson": { "all": 0.6090786981125251, "mean": 0.629903775298695, "wmean": 0.6451270836208292 }, "spearman": { "all": 0.5915463354166586, "mean": 0.6487767164109794, "wmean": 0.6539936080030831 } } }, "STS13": { "FNWN": { "pearson": [ 0.5827724320641434, 1.399853624093483e-18 ], "spearman": [ 0.6243224304995846, 8.1205413287147755e-22 ], "nsamples": 189 }, "headlines": { "pearson": [ 0.782141211640678, 7.962032809197313e-156 ], "spearman": [ 0.7864730688106963, 1.0734436155388987e-158 ], "nsamples": 750 }, "OnWN": { "pearson": [ 0.7763285115488442, 3.939259823816713e-114 ], "spearman": [ 0.7706743357210359, 1.7455185991511694e-111 ], "nsamples": 561 }, "all": { "pearson": { "all": 0.7269930269261002, "mean": 0.7137473850845552, "wmean": 0.7548467955796887 }, "spearman": { "all": 0.7392805868780532, "mean": 0.7271566116771057, "wmean": 0.7601333622079632 } } }, "STS14": { "deft-forum": { "pearson": [ 0.5228897751421426, 6.073682402382636e-33 ], "spearman": [ 0.5053330319071232, 1.532862819818703e-30 ], "nsamples": 450 }, "deft-news": { "pearson": [ 0.7680069983657648, 1.2852042740473546e-59 ], "spearman": [ 0.7189884605027314, 5.271976030712246e-49 ], "nsamples": 300 }, "headlines": { "pearson": [ 0.7541282525177266, 9.793324818317855e-139 ], "spearman": [ 0.7184718603499471, 4.702376618230486e-120 ], "nsamples": 750 }, "images": { "pearson": [ 0.8478349220740108, 2.8453760358482384e-208 ], "spearman": [ 0.8081484798616755, 3.9047867377054607e-174 ], "nsamples": 750 }, "OnWN": { "pearson": [ 0.8215837025605404, 9.871899662708195e-185 ], "spearman": [ 0.8279072975227956, 4.92735009919912e-190 ], "nsamples": 750 }, "tweet-news": { "pearson": [ 0.7317126218269044, 1.2202543536716308e-126 ], "spearman": [ 0.6520317484632859, 5.3026345548268765e-92 ], "nsamples": 750 }, "all": { "pearson": { "all": 0.7189696594882792, "mean": 0.7410260454145149, "wmean": 0.7552392326821546 }, "spearman": { "all": 0.670800547509326, "mean": 0.7051468131012597, "wmean": 0.7194709179086142 } } }, "STS15": { "answers-forums": { "pearson": [ 0.7262407045308339, 1.1025031075653589e-62 ], "spearman": [ 0.7275038094083748, 5.324790051370908e-63 ], "nsamples": 375 }, "answers-students": { "pearson": [ 0.6978521677757564, 1.6036590447498983e-110 ], "spearman": [ 0.7024365108998977, 1.4353461268590496e-112 ], "nsamples": 750 }, "belief": { "pearson": [ 0.7839138106441258, 3.235687910099277e-79 ], "spearman": [ 0.79336254174086, 2.047534842473332e-82 ], "nsamples": 375 }, "headlines": { "pearson": [ 0.8124656869189333, 1.9047661503782033e-177 ], "spearman": [ 0.8124179763234277, 2.0744785143960583e-177 ], "nsamples": 750 }, "images": { "pearson": [ 0.8731830765774482, 1.1398312421407709e-235 ], "spearman": [ 0.8795017881810903, 1.9914488519259588e-243 ], "nsamples": 750 }, "all": { "pearson": { "all": 0.7827019127153672, "mean": 0.7787310892894196, "wmean": 0.7846445472149044 }, "spearman": { "all": 0.790296617367925, "mean": 0.7830445253107301, "wmean": 0.7886973627447582 } } }, "STS16": { "answer-answer": { "pearson": [ 0.7325081931498698, 5.607126931548686e-44 ], "spearman": [ 0.7313808706110454, 8.788498983014453e-44 ], "nsamples": 254 }, "headlines": { "pearson": [ 0.7702918518354713, 3.5986232616752304e-50 ], "spearman": [ 0.7838189633800285, 5.030858666324319e-53 ], "nsamples": 249 }, "plagiarism": { "pearson": [ 0.8193777086614761, 5.14320199251657e-57 ], "spearman": [ 0.8274487763356487, 4.601518050831221e-59 ], "nsamples": 230 }, "postediting": { "pearson": [ 0.8373109699109689, 2.1202879642927884e-65 ], "spearman": [ 0.8554383865248261, 4.218785247850647e-71 ], "nsamples": 244 }, "question-question": { "pearson": [ 0.7703959894794563, 2.4658565942110764e-42 ], "spearman": [ 0.7772827176027448, 1.5625377662687375e-43 ], "nsamples": 209 }, "all": { "pearson": { "all": 0.7784793728977035, "mean": 0.7859769426074485, "wmean": 0.785525517385094 }, "spearman": { "all": 0.7906179855205325, "mean": 0.7950739428908588, "wmean": 0.7946323236636283 } } }, "eval_senteval-STS12": 0.5915463354166586, "eval_senteval-STS13": 0.7392805868780532, "eval_senteval-STS14": 0.670800547509326, "eval_senteval-STS15": 0.790296617367925, "eval_senteval-STS16": 0.7906179855205325, "eval_senteval-STSBenchmark": 0.7534738635333297, "eval_senteval-SICKRelatedness": 0.6909477744903625, "eval_senteval-avg_sts_7": 0.7181376729594552, "eval_senteval-MR": 81.19, "eval_senteval-CR": 87.12, "eval_senteval-SUBJ": 95.61, "eval_senteval-MPQA": 89.04, "eval_senteval-SST2": 85.21, "eval_senteval-TREC": 80.69, "eval_senteval-MRPC": 71.69, "eval_senteval-avg_transfer": 84.36428571428571 }