FewCLUE小样本学习榜(提交多份) Github地址 | 提交样例 | 测评方案 | NLPCC2021-FewCLUE决赛成绩 | 测评规则(新)
模型描述需包含关键词"FewCLUE.M"; 提交需实名，即：队伍名称、模型名称、Url/Github、模型描述，需有真实有效。无意义的提交将被移除；有问题发邮件：CLUEbenchmark@163.com
2021-06-08: 小样本榜-多份提交启用；

排行	模型	研究机构	测评时间	Score	EPRSTMT	CSLDCP	TNEWSF	IFLYTEKF	OCNLIF	BUSTM	CHIDF	CSLF	CLUEWSCF
1	Human	CLUE	21-05-07	83.934	90.0(0,0)	68.0(0,0)	71.0(0,0)	66.0(0,0)	90.3(0,0)	88.0(0,0)	87.1(0,0)	84.0(0,0)	98.0(0,0)
2	RobustPrompt	微信 AI	21-10-13	69.321	87.44(1.07,88.7)	59.02(0.82,65.3)	74.05(1.03,77.2)	44.42(0.82,52.4)	71.04(0.61,71.7)	75.23(2.79,79.2)	68.35(0.94,71.1)	76.37(0.37,77.2)	72.62(1.85,72.1)
3	pt_test9_unlimited	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-29	66.128	85.84(0.9,87.8)	60.78(1.16,65.5)	71.67(3.06,74.3)	46.07(0.71,51.8)	71.86(0.58,73.1)	69.21(3.66,76.8)	72.51(0.94,74.6)	65.85(5.2,69.4)	56.83(1.02,67.6)
4	pt_test9	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-29	65.334	85.92(1.38,87.9)	57.6(0.94,64)	72.69(1.44,77.1)	44.78(1,52.1)	70.72(0.52,70.5)	61.45(5.47,72.4)	73.83(1.54,75.6)	68.37(2.87,73.5)	59.93(6.71,68.3)
5	test_sparse	皮皮虾(长虹AI实验室)	21-06-29	64.525	85.5(0.93,85.9)	59.17(0.8,63.5)	72.84(3.68,72.6)	44.04(1.12,52)	67.77(1.01,70.3)	74.16(1.64,76.8)	58.05(1.75,62.6)	60.75(3.7,73.2)	66.69(2.49,62.1)
6	test	皮皮虾(长虹AI实验室)	21-06-29	64.344	85.5(0.93,85.9)	59.17(0.8,63.5)	72.84(3.68,72.6)	44.04(1.12,52)	68.21(1.27,70.9)	73.03(1.06,76)	58.05(1.75,62.6)	60.75(3.7,73.2)	65.93(1.77,69.3)
7	test２	皮皮虾(长虹AI实验室)	21-07-04	64.114	84.94(0.93,85.4)	59.43(1.02,63.3)	72.23(4.37,74.5)	41.6(1,44.9)	68.02(1.85,70.8)	75.63(0.57,77.6)	58.47(0.94,61)	60.41(3.58,69.1)	64.34(4.23,71.7)
8	unlimited_track_pt_5	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-27	63.527	85.82(1.47,86.6)	60.37(1.4,65.7)	73.79(1.65,79.1)	47.37(0.83,54.6)	72.48(0.47,72.5)	70.27(1.82,75)	47.74(2.01,62.4)	66.57(4.27,73.3)	57.52(3.43,65.5)
9	unlimited_track_pt_6	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-27	63.369	85.82(1.47,86.6)	60.37(1.4,65.7)	73.79(1.65,79.1)	47.37(0.83,54.6)	72.48(0.47,72.5)	70.27(1.82,75)	46.48(4.03,64.8)	66.57(4.27,73.3)	57.52(3.43,65.5)
10	Fewclue_mmpt1	姜汁柠檬(腾讯云小微教育)	21-06-29	63.112	87.73(1.15,88)	60.26(0.97,65.6)	73.07(1.17,74.5)	45.25(2.01,48.8)	66.18(1.83,71)	70.4(2.51,78.4)	57.3(2.8,67.4)	56.94(3.83,68.3)	60.76(6.85,67.2)
11	Fewclue_mmpt	姜汁柠檬(腾讯云小微教育)	21-06-29	62.802	87.73(1.15,88)	60.26(0.97,65.6)	73.13(0.83,74.7)	45.25(2.01,48.8)	65.93(2.1,71)	70.4(2.51,78.4)	57.3(2.8,67.4)	54.71(2.11,67.7)	60.76(6.85,67.2)
12	limited_track_pt_8	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-28	62.739	87.28(1.03,87.3)	57.73(0.56,64.8)	74.21(1.52,72.3)	45.35(0.79,52.1)	70.55(0.69,71.6)	64.45(8,73.4)	47.29(3.74,59.8)	65.74(3.06,73.4)	63.45(7.28,72.8)
13	limited_track_pt_4	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-27	62.128	85.5(2.46,87.5)	57.66(0.65,62.8)	71.2(2.58,70.5)	45.24(1.24,48)	70.55(0.69,71.6)	64.45(8,73.4)	46.78(2.68,64.4)	65.74(3.06,73.4)	61.03(6.86,66.2)
14	unlimited_track_ad_pt_3	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-24	61.818	86(1.9,71.7)	60.09(1.26,24.3)	75.29(3.88,32.9)	46.08(0.82,27.6)	71.58(0.51,34.4)	65.7(6.14,58.8)	40.45(4.87,14.4)	63.95(2.07,73.2)	60.62(2.24,62.4)
15	limited_track_pt_6	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-27	61.776	87.28(1.03,87.3)	57.73(0.56,64.8)	74.21(1.52,72.3)	45.35(0.79,52.1)	70.25(0.45,72.8)	59.23(6.22,74.2)	47.29(3.74,59.8)	63.55(3.04,70.7)	63.45(7.28,72.8)
16	limited_track_pt_7	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-28	61.348	87.28(1.03,87.3)	56.98(1.16,63.6)	72.81(1.71,72.9)	42.78(0.87,48.8)	68.16(2.73,71)	62.39(5.97,72.6)	46.57(2.76,62.1)	65.52(2.73,73.1)	61.03(7.15,72.8)
17	Fewclue_mpt_s	姜汁柠檬(腾讯云小微教育)	21-06-25	61.170	87.07(1.46,87.4)	60.37(0.86,65.6)	73.17(1.6,3.8)	47.96(0.55,51.7)	64.43(3.78,70.5)	65.48(1.72,75.4)	58.29(1.87,68.1)	50.59(0.82,62.4)	55.1(11.47,67.2)
18	Fewclue_mpt	姜汁柠檬(腾讯云小微教育)	21-06-25	61.154	87.04(1.17,87.1)	60.37(0.86,65.6)	73.31(1.73,75.9)	47.96(0.55,51.7)	64.43(3.78,70.5)	65.48(1.72,75.4)	58.29(1.87,68.1)	50.49(0.61,63.4)	55.1(11.47,67.2)
19	ptpet_self_train	MLP fans(百度研究院商业智能实验室)	21-06-29	60.525	86.85(0.4,85.5)	59.95(1.34,66.8)	71.23(1.53,73.5)	45.34(2.95,48.2)	69.25(1.33,70.8)	65.42(1.6,68.9)	37.45(2.1,41.7)	64.77(0.21,67.8)	55.1(6.14,61.7)
20	limited_track_pt_3	篮网总冠军(阿里巴巴达摩院&计算平台PAI)	21-06-25	60.328	85.71(1.25,87.1)	57.5(0.92,62.7)	73.2(1.47,68.8)	44.68(0.67,51.1)	68.57(1.72,70.1)	64.89(3.83,71.7)	43.04(3.11,56.2)	58.5(4.25,73.8)	59.66(9.01,63.1)

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

	C	E	N
C	0.0	0.0	0.0
E	0.0	0.0	0.0
N	0.0	0.0	0.0

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