Editing Factual Knowledge in Language Models

This paper proposes to train a hyper-network to patch the factual knowledge in PLMs. Specifically, they train a BiLSTM to predict the update to a PLM given the sentences to be corrected and another set of sentences that are not intended to be modified. Their training objective is a vanilla log likelihood constrained by a condition that the KL divergence of distribution of non-modified sentence pairs before and after updates is less than a margin.

Comments

  • This hyper-net concept is quite new to me and their design seems to be very interesting. However, I can tell this won’t be so popular due to computational costs.
Rating
  • 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.

0 投票人