This NeurIPS 2021 paper presents a simple ensembling method to ensemble graph prediction and achieved success of +1.x points on AMR parsing. Their method involves matching multiple graphs using the hill-climbing algorithm and counting how many times an edge or a vertex is matched. This frequency is used as a vote to filter out low confident sub-predictions.
- Direction is good. I never thought about exploring an ensembling algorithm for parsing.
- The algorithm is purely heuristic.
- Though the algorithm is simple, they spent quite a whole a page to describe it, which frightened me the first time I read it. I would appreciate it if they would had written a shorter version.
- I have a question though. What happens if an edge gets a high vote but one of its vertex gets very low vote?
- 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.