Multi-Task Learning on MGSM

Metric: Average (%) (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	Average (%)▼	Augmentations	Paper	Date↕	Code
1	PaLM 2 (few-shot, k=8, SC)	87	No	PaLM 2 Technical Report	2023-05-17	Code
2	PaLM 2 (8-shot, CoT)	72.2	No	PaLM 2 Technical Report	2023-05-17	Code
3	Flan-PaLM 540B (8-shot, fine-tuned, CoT + SC)	72	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
4	Flan-U-PaLM 540B (CoT)	60.4	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
5	Flan-PaLM 540B (8-shot, fine-tuned, CoT)	57	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
6	PaLM 540B	55	No	PaLM: Scaling Language Modeling with Pathways	2022-04-05	Code
7	U-PaLM 540B (CoT)	49.9	No	Transcending Scaling Laws with 0.1% Extra Compute	2022-10-20	-
8	text-davinci-003	36	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
9	code-davinci-002	35	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
10	text-davinci-002	23.7	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
11	Flan-PaLM 540B (8-shot, fine-tuned)	21.2	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code
12	GPT-3 Davinci 175B	5.7	No	Scaling Instruction-Finetuned Language Models	2022-10-20	Code

#1PaLM 2 (few-shot, k=8, SC)SOTA
87
Average (%)· 2023-05-17
PaLM 2 Technical Report Code
#2PaLM 2 (8-shot, CoT)
72.2
Average (%)· 2023-05-17
PaLM 2 Technical Report Code
#3Flan-PaLM 540B (8-shot, fine-tuned, CoT + SC)SOTA
72
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#4Flan-U-PaLM 540B (CoT)
60.4
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#5Flan-PaLM 540B (8-shot, fine-tuned, CoT)
57
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#6PaLM 540BSOTA
55
Average (%)· 2022-04-05
PaLM: Scaling Language Modeling with Pathways Code
#7U-PaLM 540B (CoT)
49.9
Average (%)· 2022-10-20
Transcending Scaling Laws with 0.1% Extra Compute
#8text-davinci-003
36
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#9code-davinci-002
35
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#10text-davinci-002
23.7
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#11Flan-PaLM 540B (8-shot, fine-tuned)
21.2
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code
#12GPT-3 Davinci 175B
5.7
Average (%)· 2022-10-20
Scaling Instruction-Finetuned Language Models Code