Neural, Neural Everywhere: Controlled Generation Meets Scaffolded,

Dialogue System
#1

This paper presents the second-place chatbot in the Alexa Prize Socialbot Grand Challenge 4. Its novelty is to partition the bot into several response generators, each of which is responsible for a given topic. Their dialog manager is also entity-centric as the winner system. They keep a list of entities that are mentioned or discussed and forward the dialog flow to generators in related domain. Each generator utilizes different neural seq2seq models, but in general a template based controlled generation and a paraphrase system are utilized.2

Comments

  • The partition design is very robust.
  • Their exploitation of neural paraphrase is so good as most facts are extracted from Wikipedia, which are of high quality. A neural paraphrase module makes these facts more conversational and causal such that the user will feel these sounds like a dialog instead of a written language.
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 voters