lfqian commited on
Commit
c12f840
·
verified ·
1 Parent(s): bd28f1d

Upload model_performance.csv

Browse files
Files changed (1) hide show
  1. model_performance.csv +18 -0
model_performance.csv ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Models,Average,FinQA,DM-Simplong,XBRL-Math,Type
2
+ GPT-4o,68.24,72.49,60.0,72.22,instruction-tuned
3
+ GPT-o1,59.84,49.07,56.0,74.44,instruction-tuned
4
+ GPT-o3-mini,65.51,60.87,59.0,76.67,instruction-tuned
5
+ DeepSeek-V3,67.62,73.2,53.0,76.67,instruction-tuned
6
+ DeepSeek-R1,68.93,65.13,53.0,86.67,instruction-tuned
7
+ Qwen2.5-72B-Instruct,66.72,73.38,59.0,67.78,instruction-tuned
8
+ Qwen2.5-72B-Instruct-Math,65.69,69.74,42.0,83.33,instruction-tuned
9
+ DeepSeek-R1-Distill-Llama-70B,68.8,66.73,53.0,86.67,instruction-tuned
10
+ Llama3-70B-Instruct,52.2,58.92,41.0,56.67,instruction-tuned
11
+ Llama3.1-70B-Instruct,58.17,63.18,48.0,63.33,instruction-tuned
12
+ Llama3.3-70B-Instruct,64.05,68.15,54.0,70.0,instruction-tuned
13
+ DeepSeek-R1-Distill-Qwen-32B,68.97,65.48,55.0,84.44,instruction-tuned
14
+ DeepSeek-R1-Distill-Qwen-14B,63.9,63.27,44.0,84.44,instruction-tuned
15
+ DeepSeek-R1-Distill-Llama-8B,53.36,45.96,33.0,81.11,instruction-tuned
16
+ Llama3-8B-Instruct,39.95,41.97,29.0,48.89,instruction-tuned
17
+ Llama3.1-8B-Instruct,50.12,54.13,34.0,62.22,instruction-tuned
18
+ Fino1-8B,61.03,60.87,40.0,82.22,instruction-tuned