CLUE1.1总排行榜     CLUE1.1提交规则  |   项目地址
CLUE1.1与CLUE1.0区别:区别与原有的CLUE1.0,CLUE1.1在部分任务启用了新的测试集,训练集和验证集保持不变;CLUE1.0保留CMNLI自然语言推理任务
2021年07月25日更新TNEWS测试集;2021年09月29日更新CLUEWSC2020测试集;2021年9月5日更新CHID、C3测试集;请重新拉取数据集,并参考“提交样例”及其README.md

排行模型研究机构测评时间Score1.1认证AFQMCTNEWS1.1IFLYTEKOCNLI_50KWSC1.1CSLCMRC2018CHID1.1C3 1.1
1玉言 网易伏羲 23-07-3187.050待认证86.4574.0467.9686.3395.7397.684.2595.95695.138
2HunYuan-NLP 1T腾讯混元AI大模型团队22-11-2686.918待认证85.1170.4467.5486.59696.287.998.84893.723
3通义-AliceMind达摩院NLP22-11-2286.685待认证84.0773.4767.4285.8794.3395.0386.899.20893.969
4HUMANCLUE19-12-0186.678已认证817180.390.3988492.487.1096.00
5CHAOSOPPO研究院融智团队22-11-0986.552待认证83.3773.2265.8186.3794.695.787.299.21793.477
6WenJinMeituan NLP22-10-2086.313待认证84.4973.0464.3886.2394.4495.6786.2598.89893.415
7OBERTOPPO小布助手22-11-0784.783待认证81.0267.756684.5391.399.9384.0597.57890.892
8HunYuan_nlp腾讯TEG22-05-1184.730待认证83.3764.0166.5885.2392.2793.8787.998.51290.831
9ShenNonG云小微AI21-12-0184.351待认证82.5765.5664.4285.9794.2191.2386.597.93290.769
10ShenZhouQQ浏览器实验室(QQ Browser Lab)21-09-1983.873待认证80.5565.3667.6586.3789.0890.9787.8597.92389.108
11MusaBertmthreads22-12-1682.889待认证86.9265.2263.8881.688.9392.983.9595.88986.708
123mp_xxlargevivo-3MP23-02-2281.413待认证77.9363.464.3182.691.887.281.0597.22787.200
13vivo-3MPvivo-3MP23-03-2681.413待认证77.9363.464.3182.691.887.281.0597.22787.200
14UniMC-DeBERTa-1.4BIDEA研究院22-11-1481.354待认证78.5561.2263.8182.5392.1591.0377.995.45489.538
15CL-BERTCL-BERT22-04-0681.288待认证82.4464.0164.8182.7788.5891.4781.9598.51277.046
16PAI-EasyNLP BERTwjn199622-09-0581.176待认证77.0561.1661.1982.887.1495.6780.497.91187.262
17Mengzi澜舟科技-创新工场21-09-1481.092待认证81.7965.1665.0882.5786.4889.8783.9595.10979.815
18MTBertmthreads22-10-2179.204待认证81.0758.6163.2379.438890.778.7591.38781.662
193mp_basevivo-3MP23-02-2279.150待认证76.8560.7862.7779.5789.8284.78095.58782.277
20nlp_largebest_unknown22-12-1979.114待认证77.1659.9362.8180.4788.8985.4379.5595.75382.031

ALBERT(Ensemble)

GitHub/模型网址:

提交日期:9月17日

分数:9月17日

更多详情:

型号说明

阿尔伯特模型集合

参数说明

单任务微调。我们从MNLI为RTE、STS和MRPC优化的模型开始

总参数:-1

共享参数:-1

诊断信息

诊断主混淆矩阵

C N E
C 182 36 40
N 81 189 116
E 17 69 374

C = 对立

N = 不包含

E = 包含

获取排行榜数据成功!