TAL Journal: Explainability of NLP models (64-3)

Instructions to the reviewers

Please rate the paper according to the criteria indicated on the evaluation page, paying attention to the sub-criteria; then give an overall score for the paper.

At the end of the evaluation, please provide detailed comments to help the editorial committee make the final decision. These comments are particularly important during the first round of reviews as they will help authors to improve the quality of their article for a possible second round.

Avoid mentioning anything in the comments that would lead the author(s) to believe that your recommendation to reject or accept the article will be followed by the editorial board since the final decision depends on various factors, particularly other reviews.

In the case of a second review, do not delete your first comments; add your new comments after a separation line.

If you should have any conflict of interest, please let us know as soon as possible.

The policy of the journal is double-blind reviewing: the identity of the reviewers is not transmitted to the authors. We ask you not to distribute the article during this evaluation phase and not to mention your quality as a reviewer in any context whatsoever.

Thank you!

 

EVALUATION CRITERIA:


-----------------------------------------------------------


DETAILED SCORES (1 is poor, 5 is excellent):

RELEVANCE (1-5):
* Does the paper fit in the topics of the journal?

CLARITY (1-5):
* Is ithe scientific content clearly presented?
* Are the discussions in the paper clear?
* Is the overall organization of the paper satisfactory (order and size of the sections)? Is the technical quality of the paper satisfactory (figures, graphs, plots, formulas, etc.)?
* Do the title, abstract and keywords provide a clear, accurate indication of the material presented?

CORRECTNESS (1-5):
* Does the paper appear to be flawed technically or methodologically? Are there enough linguistic examples? Are they detailed?
* How important is the work?
* Are the techniques evaluated on sufficiently large data?
* Are the references complete and accurate?

ORIGINALITY (1-5):
* How novel is the approach? Do you consider the content of the paper to be of a high originality?

------------------------------------------------------------------

FINAL RECOMMENDATION (FIRST REVIEWING ROUND)

5 (= Accept the submission as is)

4 (= Revisions required: The paper is acceptable but requires some changes suggested in the COMMENTS TO AUTHORS section)

3 (= Resubmit for review: The paper requires significant revisions suggested in the COMMENTS TO AUTHORS section before a decision can be made. The author(s) will have to send a revised version for evaluation)

2 (= Reject. Encourage the author(s) to try a major revision or suggest to submit to another journal (specify which one))

1 (= Reject. Do not encourage another submission)

------------------------------------------------------------------

FINAL RECOMMENDATION (SECOND REVIEWING ROUND)

Publish

Reject

------------------------------------------------------------------

COMMENTS TO AUTHORS (WILL BE SEEN BY THE AUTHORS AND THE EDITORIAL BOARD):
(provide detailed and constructive comments so that authors can improve their submission)


-----------------------------------------------

COMMENTS TO THE EDITORIAL BOARD (WILL NOT BE SEEN BY AUTHORS):

Online user: 2 Privacy
Loading...