Integrating Semantic and Structural Information with Graph Convolutional...

Text classification often fails on controversial posts detection due to its weakness in capturing graph structures of posts and comments. This paper proposed the DTPC-GCN to model post-comment graphs to overcome this limitation.

Their base model is a conventional GCN encoder operating topic/post/comment nodes. Nothing new.

To disentangle child representation from topics, they proposed a novel two-branch model. The first branch learns topic unrelated features by maximizing cross entropy, while the second branch learns related features by minimizing it as usual. Finally, an attention module fuses features from these two branches to make the final prediction.


To me, the maximizing crossentropy trick is quite smart. However, the experiments are not convincing. How can BERT outperform CNN by only 1 point? The author claimed:

We only fine-tune the last layer, namely layer 11 of BERT for simplicity

which intentionally lowered down the baseline. I don’t like this one.

  • 5: Transformative: This paper is likely to change our field. It should be considered for a best paper award.
  • 4.5: Exciting: It changed my thinking on this topic. I would fight for it to be accepted.
  • 4: Strong: I learned a lot from it. I would like to see it accepted.
  • 3.5: Leaning positive: It can be accepted more or less in its current form. However, the work it describes is not particularly exciting and/or inspiring, so it will not be a big loss if people don’t see it in this conference.
  • 3: Ambivalent: It has merits (e.g., it reports state-of-the-art results, the idea is nice), but there are key weaknesses (e.g., I didn’t learn much from it, evaluation is not convincing, it describes incremental work). I believe it can significantly benefit from another round of revision, but I won’t object to accepting it if my co-reviewers are willing to champion it.
  • 2.5: Leaning negative: I am leaning towards rejection, but I can be persuaded if my co-reviewers think otherwise.
  • 2: Mediocre: I would rather not see it in the conference.
  • 1.5: Weak: I am pretty confident that it should be rejected.
  • 1: Poor: I would fight to have it rejected.