Questions about Math-Shepherd and DeepSeek-Math Paper
Hi Peiyi Wang! Congratulations for your and your fellow authors work,
I would be glad if you could explain to me your automatic process annotation framework, how do you annotate this sample - "If Buzz bought a pizza with 78 slices at a restaurant and then decided to share it with the waiter in the ratio of 5:8, with Buzz's ratio being 5, what's twenty less the number of slices of pizza that the waiter ate? Step 1: The total ratio representing the pizza is 5+8 = <<5+8=13>>13. + Step 2: The waiter ate 13 x 8 / 13 = <<13*8/13=6>>6 slices of the pizza. - Step 3: Buzz ate 78 - 6 = <<78-6=72>>72 slices of the pizza. - Step 4: The waiter ate 20 less than the number of slices that Buzz ate which is 72 - 20 = 52. - Step 5: The waiter ate 52 slices of the pizza. The answer is: 52 -", I guess this was a reasoning process output from Llemma-7b and you used Llama-2-70b to calculate the quality but how?
Like for computing the quality of S3, do you guys feed Q, S1, S2 and S3 to the completer or Q and S3? Which one do you do in your paper?
Hoping for a positive response,
Best Sen