CGO SC Meeting 6 June 2025 - cgo-conference/public GitHub Wiki

CGO Update:
"it is quite likely that CGO will be earlier next year (31 January - 4 February)”
No issue with the venue, waiting for IEEE.

Reviewer Awards (Fabrice)
I would like to discuss the opportunity to implement this year reviewer awards. I am afraid that if we don’t take time to discuss it and decide how to implement it, we will postpone it again and again.

My idea is to provide two levels: the first level would be awarded to anybody who provide reviews on time; participate to discussions for their own assigned papers; are positively rated by other reviewers (note that review ratings should I think be double blind and that criteria of a good review should be provided). The second level would be for a few reviewers who are above the others eg either because of the way they get involved in all discussions (not only for their paper) or because their expertise makes a big difference. I would make the awards public on the conference web page.

  • Two kinds of awards:
    All reviewers who did their job get the base award
    Exceptional reviewers get an exceptional award

  • Special committee for reviewer awards?

    • Have grad students look at reviews (anonymously) and rate them?

  • Maintain database of likely reviewers and possibly not so good reviewers (with a fixed time limit of maintaining comments on reviewer's engagement)

  • Include topic of interest in this database

  • SIGPLAN had this, but it doesn't seem to have been successful (?)

  • Look at ASPLOS distinguished reviewer

  • Release reviews anonymously afterwards?

  • Open Review https://openreview.net

  • Spotting distinguished reviewers should be easy

  • Ask authors for feedback?

  • Distinguished reviewer award:

    • PC chairs + one SC member meet and decide on distinguished reviewer

Uniform Online Review Form (Xingling)
Adopt the same online review form each year as done by many similar conferences to streamline the review process. This needs to be strictly enforced.

Consistent Paper Award Policies (Xingling)
Future chairs should adhere to policies for best paper and ToT paper selection unless approved by the SC. For example, this year’s distinguished paper awards were decided by online voting among PC members (despite being informed of our practice)—an approach I find problematic since only a small fraction voted, often relying solely on scores without reading reviews. Our policy is to form a small committee to select these awards. This year, I chaired the ToT paper selection committee, and we discussed all candidates thoroughly before reaching consensus. Additionally, we should aim to award Distinguished Paper Awards to the top 10% of accepted papers each year; this year, only 3 awards were given, though 5 could have been awarded. These awards recognize the top 10% of papers for the year and are not intended as stepping stones to Turing prizes.

Paper Classifications (Xingling)
Authors should indicate the paper category (research, tool, or experience) both in the online form and immediately after the title. Some PC members default to treating every submission as a research paper and, if they rate it poorly, often refuse to adjust their scores even after the rebuttal process.

CGO Hall of Fame (Xingling)
Establish and maintain a Hall of Fame for CGO, with a minimum number of papers required for inclusion.

To better promote our conference, I propose once again that we make this official.

The current version is outdated and unofficial: https://liberty.princeton.edu/Fame/CGO/

Negative Results (Fabrice)

There is also my idea of having a mechanism to label reports (no acm publication) of negative results that seemed to interest some members of the steering committee. This is less urgent but we could discuss this matter.

  • Not a full publication, not reviewed, but maybe artifact evaluated

  • Link on the webpage to arxiv papers that might be interested to the community

  • Possibly student committee / artifact evaluation committee evaluating it

  • A workshop would be a possibility? Include poster session?

    "Our goal is to find papers which the community can learn from and might otherwise have trouble finding a suitable venue, so we take a broad view of what constitutes a "negative" result."

    https://nope.pub/

Repeated Review (Ayal)

  • Should we treat reviewers re-reviewing the same paper differently?
  • Should reviewers flag if they re-review a paper (and we assign new reviewers)
  • Should authors provide information how their paper has changed since the prior submission (in a time frame?)
    • Do we do this always, do we do this on demand?
  • Careful not to bias against resubmitted papers
  • Authors could optionally submit information about what has changed
    • This is only available to the PC chairs and is given to reviewers only on demand