Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / CLUEbenchmark/SuperCLUE issues and pull requests
#50 - 收钱瞎评的傻x山寨公司,早点倒闭
Issue -
State: open - Opened by Wangwentao1 6 days ago
#49 - 请求对最新的deepseek-r1模型进行测评
Issue -
State: open - Opened by g1021895463 29 days ago
#48 - 断更了???
Issue -
State: open - Opened by yiliangfang 3 months ago
- 1 comment
#47 - 请问评测的原理是什么呀,是人工打分的吗,如果是客观题是直接比较返回的答案的字符串,主观题是人工评判答案吗
Issue -
State: open - Opened by starplatinum3 7 months ago
- 2 comments
#46 - 工具的评测是什么含义? 是Function calling么,如果没有请添加此能力的评测。
Issue -
State: open - Opened by goqw 10 months ago
- 1 comment
#45 - Claude3有评测吗?
Issue -
State: open - Opened by Pancat007 11 months ago
#44 - 请问如何对自己做的大模型使用该指标进行测评呢?
Issue -
State: open - Opened by AWangji 11 months ago
- 4 comments
#43 - What are expected to submit for the leaderboard integration?
Issue -
State: closed - Opened by zhimin-z 12 months ago
- 1 comment
#42 - 请问一月榜单呢
Issue -
State: open - Opened by kindle939393 about 1 year ago
- 2 comments
#41 - GPT4-Turbo is missing from the general leaderboard
Issue -
State: closed - Opened by zhimin-z about 1 year ago
- 1 comment
#40 - 咨询一下,从测评报告来看,SuperCLUE是采用自动化方式的客观评估,是否可提供针对某一模型的可实际运行的自动化评测的python样例代码(api调用或者web)?
Issue -
State: open - Opened by Romanzhang2024 about 1 year ago
#39 - Does it indicate using 5 shots for evaluation?
Issue -
State: closed - Opened by zhimin-z about 1 year ago
- 1 comment
#38 - Where to download the benchmark dataset?
Issue -
State: open - Opened by zhimin-z about 1 year ago
#37 - How to calculate the metrics from the table in the paper to the leaderboard?
Issue -
State: open - Opened by zhimin-z about 1 year ago
- 1 comment
#36 - 大模型升级方式
Issue -
State: open - Opened by lukeup about 1 year ago
#35 - 想问下 角色扮演 benchmark是怎么进行的
Issue -
State: closed - Opened by xealml about 1 year ago
#34 - Where to locate the SuperCLUE-LYB leaderboards?
Issue -
State: open - Opened by zhimin-z over 1 year ago
#33 - 能否增加翻译的评估排名
Issue -
State: open - Opened by lx0126z over 1 year ago
#32 - 任务规划和工具使用的评价标准是什么样的?
Issue -
State: open - Opened by heibaidaolx123 over 1 year ago
- 1 comment
#31 - c-eval是真的离谱,希望superclue能更新的稍微快一点,比如1-2周更新一次
Issue -
State: open - Opened by iammeizu over 1 year ago
- 2 comments
#30 - anthropic拼错了
Issue -
State: open - Opened by JerryJiang12923 over 1 year ago
#29 - 求教一下 逻辑与推理 具体指哪方面? 比如 "郭德纲2岁会看报,xxxx" ,请问郭德纲3岁会看书吗? 这个属于推理还是语义理解能力??
Issue -
State: open - Opened by ArtificialZeng over 1 year ago
#28 - 请问可以把vicuna-33B模型加入评测吗?
Issue -
State: open - Opened by Mr-wang2016 over 1 year ago
#27 - 测评时如何与标准答案进行匹配
Issue -
State: open - Opened by Starry-Hu over 1 year ago
#26 - 数据集开源吗?可以在哪里下载呢
Issue -
State: open - Opened by vanshaw2017 over 1 year ago
- 3 comments
#25 - 关于prompt设计的问题
Issue -
State: open - Opened by lrs1353281004 over 1 year ago
- 1 comment
#24 - 排名变化的原因是什么?
Issue -
State: open - Opened by zhaojiawen-coding over 1 year ago
- 1 comment
#23 - test the 智源大模型吧
Issue -
State: open - Opened by forkyguo over 1 year ago
- 3 comments
#22 - 阿里的通义千问没有吗?
Issue -
State: closed - Opened by Pancat009 over 1 year ago
- 2 comments
#21 - 这里"idea-jiangzhiya"应该是"idea-jiangziya"吧?
Issue -
State: open - Opened by ilongshan over 1 year ago
- 1 comment
#20 - 没有文心一言吗
Issue -
State: closed - Opened by p81sunshine over 1 year ago
- 1 comment
#19 - Clarify which "Claude" is benchmarked?
Issue -
State: open - Opened by jekbradbury over 1 year ago
- 1 comment
#18 - 可以在superclue上测试自己的模型吗?
Issue -
State: open - Opened by guozhiyao almost 2 years ago
- 2 comments
#17 - 什么时候回公开测试数据集?
Issue -
State: open - Opened by wangrui6 almost 2 years ago
- 1 comment
#16 - 建议补全人类的“专业能力”数据
Issue -
State: open - Opened by Triang-jyed-driung almost 2 years ago
- 1 comment
#15 - 人类的数值怎么来的?
Issue -
State: closed - Opened by So0ni almost 2 years ago
- 3 comments
#14 - 置信度
Issue -
State: closed - Opened by littlepan0413 almost 2 years ago
- 1 comment
#13 - 公开评测集和评测标准
Issue -
State: closed - Opened by plmsmile almost 2 years ago
- 2 comments
#12 - 开始搞手机测评榜那一套了?GPT4对应苹果,国产大模型对应华米OV
Issue -
State: open - Opened by ZhuGeRoastedFish almost 2 years ago
- 3 comments
#11 - 测评结果为什么全是整数?
Issue -
State: open - Opened by ltz0120 almost 2 years ago
- 1 comment
#10 - 这个评测的参考价值
Issue -
State: closed - Opened by liuyajun52 almost 2 years ago
- 2 comments
#9 - 作为一个测评榜,建议参考Chinese-LLaMA-Alpaca进行适度的测评说明和公开
Issue -
State: open - Opened by shm007g almost 2 years ago
- 1 comment
#8 - 评测数据客观公正很重要
Issue -
State: open - Opened by shichengustc almost 2 years ago
- 3 comments
#7 - 单项能力有多少道题目啊
Issue -
State: open - Opened by leonall almost 2 years ago
- 2 comments
#6 - 这个superCLUE 有毒性和偏见等方面的评测吗
Issue -
State: open - Opened by devinbai almost 2 years ago
- 3 comments
#5 - 生成与创作如何用选择题的形式测试的?
Issue -
State: closed - Opened by Howardqlz almost 2 years ago
- 4 comments
#4 - 该如何引用你们的工作?
Issue -
State: closed - Opened by MikeGu721 almost 2 years ago
- 1 comment
#3 - 我个人使用后的感受,星火大模型是真的不如文心一言。。
Issue -
State: open - Opened by MysteryMulberry almost 2 years ago
- 8 comments
#2 - 超200人了,求拉群
Issue -
State: closed - Opened by dinglei8908 almost 2 years ago
- 1 comment
#1 - 感谢徐亮老师团队的工作~关于评测细节 有一些疑问咨询下
Issue -
State: open - Opened by lrs1353281004 almost 2 years ago
- 5 comments