search

salesforce / ctrl-sum

Resources for the "CTRLsum: Towards Generic Controllable Text Summarization" paper
https://arxiv.org/abs/2012.04281
BSD 3-Clause "New" or "Revised" License
146 stars 24 forks source link

Question about the baselines used in the paper #12

Closed rigggby closed 2 years ago

rigggby commented 2 years ago

❓ Questions and Help

Before asking:

  1. search the issues.
  2. search the docs.

What is your question?

Hi, thanks for sharing your great work! I have some question regarding the baseline result in the paper, specifically in Table 6, is the BART here finetuned on the arXiv dataset like your CTRLsum? Also, is the result in this table all zero-shot (you did not train on these two specific tasks)? Thanks.

Code

#### What have you tried? #### What's your environment? - fairseq Version (e.g., 1.0 or master): - PyTorch Version (e.g., 1.0) - OS (e.g., Linux): - How you installed fairseq (`pip`, source): - Build command you used (if compiling from source): - Python version: - CUDA/cuDNN version: - GPU models and configuration: - Any other relevant information:
jxhe commented 2 years ago

Hi,

Thank you for your interest!

  1. Yes, the BART baseline in Table 6 is finetuned on the arXiv dataset as well
  2. Yes, the results in Table 6 are zero-shot
rigggby commented 2 years ago

Hi, thanks for your answer and have a nice day!