How We Check Reviewer Eligibility for CHI 2027

Authors: Anna Cox, Tony Tang, Thomas Kosch, Erin Solovey, Petra Isenberg, Regan Mandryk
Date: 2026-09-09

Good news! We know that some authors are struggling to get DBLP into author profiles ahead of the deadline. We will be able to provide a few days grace on this. Please make sure you and your coauthors fix this by Monday, 14 Sept AOE.

CHI 2027 asks every submission to commit four review-responsibility slots to authors who are qualified to review, and at least one to a senior author, if the paper has one. The qualification bar is the SIGCHI policy on minimum qualifications for full paper peer review roles. With roughly 9,000 submissions expected, we automatically check them against DBLP. In the following, we explain the steps taken, the information authors need to provide, which checks occur automatically, which require manual review, and when authors are queried to provide more information.

Our tools answer two questions

We need to ensure that the CHI Full Paper Review Responsibility Policy is fulfilled. As such, we need to ensure that eligible authors fulfill their responsibilities and that non-eligible authors are exempt. For performing these checks, we need to answer two questions for each submitting author:

• Who is this person? We match the person on your PCS author list to a DBLP profile.
• Is the person eligible to review? We count the full papers on that profile in CHI and the SIGCHI specialized conferences, and assess if they are eligible under the policy for a minimum qualification for full paper peer review roles.

We only apply a label (see labels at bottom of this post) when we are confident about the identity.

Who is this person?: Working out who you are

To identify each author, we follow the steps outlined next:

1. A DBLP URL in your PCS profile. You told us which page is yours. Nothing overrides this except a chair looking at the record by hand.
2. Your ORCID. DBLP tags many publications with author ORCIDs, and dblp.org/orcid/<id> resolves to a profile. If neither works, we ask ORCID for your own public works list and find those papers’ DBLP records by DOI, which identifies you through a shared publication.
3. Your name. This is helpful to disambiguate between people; however, with common names, it is error-prone on its own.

Sometimes it is difficult to identify individuals. For example:

(1) Name collisions, and inaccurate DBLP records. We are finding this happens quite a bit for: (a) junior students, and (b) people with names that have been Anglicized

– e.g. Elias Müller has a DBLP entry, but it is a mix of at least two different people (working with the same supervisor!)

– e.g. many Chinese scholars collide into the same name; and, many of these scholars don’t use DBLP as “single source of truth” for their publication records

(2) People that put in N/A

– e.g. Anthony Tang has several entries (numbered), but if Anthony Tang entered in N/A, we cannot guess which one is his, or if it is true that he is actually N/A. 

If we cannot identify you, then we will contact you to clarify.

Why are we not performing string name matching?

DBLP handles shared names by numbering people — Wei Wang 0001 through Wei Wang 0520 — while also keeping a plain Wei Wang entry that holds every publication not yet attributed to a specific person. Looking up a name matches that plain entry, which describes nobody in particular. In our own data, one such entry held a single publication (making a researcher with 146 look like they had none), and another held 1,835 papers by many different people.

We resolve names against the whole family of same-named profiles and split them using your affiliation. When the family cannot be split, the answer is “ambiguous,” and a human looks at it.

Another challenge arises if you are a new author: an established author may also share your name. Thus, we do not match against DBLP with the name alone.

“What if I have no DBLP page”?

You can say so in PCS (enter “N/A” in PCS), and we take it seriously, though our tools do check. Of the people who declared no DBLP page in one recent snapshot, about a quarter turned out to have one. So we record “confirmed: no DBLP record” only when you declared it, our lookups found nothing, and ORCID’s own list of your works matched no indexed publication.

What if my co-author has not put in their DBLP?

You and your co-author will receive emails reminding you to fix this. We are able to provide a few days grace to get this fixed after the submission deadline – until Monday 14th Sept AOE. 

If an author fails to update their profile and add a DBLP or write N/A then the burden for fixing this would fall on the volunteers. And it’s just not scale-able. Therefore, if you do not have this filled in you will not be able to submit.

Is this person eligible to review?: Counting the record

Based on the review policy, we use three nested sets of venues:

• CHI — the CHI main conference.
• Eligible venues — CHI plus all SIGCHI specialized conferences, see the full list here https://sigchi.org/ , namely: AutoUI, C&C, CHI Play, CI, COMPASS, CSCW, CUI, CritiCo, DIS, EICS, ETRA, EduCHI, GROUP, HRI, ICMI, IDC, IH, IMX, ISS, IUI, MobileHCI, RecSys, SCG, SUI, TEI, UIST, UMAP, UbiComp / ISWC, VRST.
• HCI venues — the eligible list plus various HCI conferences and journals.

A publication counts as a full paper when it is not an arXiv preprint or a withdrawn record, not in an Extended Abstracts / Adjunct / Companion volume, and long enough: 4 pages at CHI, 7 pages elsewhere. CHI’s lower bar is deliberate — CHI Notes (roughly 2003–2015) were peer-reviewed main-proceedings papers capped at four pages, and DBLP records them the same way as full-length papers.

We use these counts to derive labels for reviewer types:

Label	Rule
AC Eligible	≥ 5 full papers in HCI venues and ≥ 2 full CHI papers
Review Eligible	≥ 1 full paper at CHI or a SIGCHI specialized conference
Ineligible	no full paper at any of those venues

Both AC Eligible and Review Eligible authors can fulfill review-responsibility slots. The AC Eligible label matters because a paper with at least one author fulfilling this criterion must be listed in the review-responsibility slots.

What you will see

If your submission’s slots do not check out, you get an email listing every author on the paper in three groups (eligible, not eligible, and identity unconfirmed) with the counts behind each verdict and a link to the DBLP page those counts came from. It also names the specific PCS slots to change. If one eligible author has to hold several slots, that is allowed, and the letter says so.

In some cases, there is nothing for authors to do. For example, if no author on the paper is qualified to review, no nomination could have satisfied the rule; we record it and close it, and you do not get an email.

If we got it wrong

Even though we try to be cautious with the categorization and use manual reviewing in cases where the system is unsure, we still might get it wrong and you will get an email. 

If this is the case, please email us telling us the correct assessment. Here are some of the likeliest causes, roughly in order:

• Your PCS profile has no DBLP URL or ORCID. This is the single fix that resolves most cases. Add either one and the next run picks it up.
• We matched the wrong person with your name. The email shows the DBLP page we counted, so this is visible immediately.
• Your papers are on a profile we did not find. A new profile, a name variant, or publications split across two DBLP entries.
• Your qualifying paper is too recent to be indexed, or DBLP files its venue under a key we do not count.

If we got it wrong, reply to the message and tell us. We can override the automatic decision directly, in either direction, and that override outranks everything the tools derive.