NeurIPS 2026’s call for papers initially set deadlines of May 4 for abstracts and May 6 for paper submissions. Initially, May 6 appears to be the significant date. That is when the paper is due. Yet without the abstract on May 4, there is no paper submission two days later.
A call for papers, or CFP, tells you what, when, and how a conference or journal wants your work.
The document can be one page or many. Somewhere inside are the topics, deadlines, submission rules, and review details. The task is to find where your paper belongs there and what you need to do before sending it.
What is a call for papers?
Consider a paper on distributed learning.
You find an artificial intelligence conference that looks promising. Then you open the CFP and find separate tracks for machine learning, distributed systems, privacy, networking, and applications.
Now the question changes. It is no longer just whether the conference fits your paper. You need to know where within the conference your paper fits.
The CFP gives you the clues. Its topics show the research the organizers want. The deadlines tell you when the paper has to arrive. The submission rules explain how to prepare them. There may also be information about peer review and what happens after acceptance.
| Element | What it tells you |
|---|---|
| Topics or tracks | The research the organizers want, and where the paper belongs |
| Deadlines | When the abstract, paper, decision, and final version are due |
| Submission rules | Template, page limit, anonymity, and supplementary files |
| Review details | How the paper will be reviewed and what happens after acceptance |
Take the formatting rules. A conference might require a specific template, eight pages, double-blind review, and a separate file for supplementary material. Another might use a different template and allow twelve pages.
The research has not changed, but the submission has to.
The same applies to the track you choose. Send a networking paper to a machine-learning track, and it may reach reviewers who are interested in a distinct problem. Miss the deadline, and it will never reach them. Leave your name on a paper submitted for double-blind review, and it may fail a basic submission check.
This is why the CFP is part of preparing the paper, not something to read after the paper is finished.
What is the difference between a call for papers and a call for abstracts?
A call for papers and a call for abstracts sound similar, but they do not always ask for the same material.
A call for abstracts starts with a brief description of the research.
Consider the annual meeting of the American Association for Cancer Research. Researchers describe the work they want to present in an abstract. Those abstracts are then used to choose talks and posters.
No full research paper is required at that stage.
A call for papers often goes further. In computer science, many conferences ask for a full manuscript, send it to reviewers, and publish accepted papers in conference proceedings.
The names alone, however, do not tell the story.
NeurIPS 2026 called for papers, but authors still had to submit an abstract first. The abstract was due on May 4 and the paper on May 6. The abstract was not an option for the paper. It was the first step toward submitting it.
Before asking whether your paper is ready, ask something simpler:
What do I actually need to submit?
How is a conference CFP different from a journal special issue?
A conference and a journal can issue a CFP. What happens after acceptance is different.
At a conference, an accepted paper becomes part of an event. You might present it as a talk or a poster. In fields such as computer science, it can also appear in conference proceedings.
The journal publishes a special issue.
Imagine a journal that publishes work from many parts of artificial intelligence. For one issue, the editors focused on distributed learning in healthcare. They invite researchers working on that subject to submit papers.
You’ll find details on the special issue’s topics, submission deadline, and the guest editors within its CFP.
The narrow topic can be useful. A study on distributed learning in hospitals, for example, already has something in common with the other papers the issue hopes to collect.
There is another question to ask, though: who is running the issue?
Some special issues have had serious problems with peer review. Wiley, for example, announced retractions involving Hindawi journals after investigations found problems with the review of papers in some special issues.
The lesson is not to avoid special issues. Established journals publish them all the time. The point is to know where the invitation comes from.
Look at the journal. Look at the publisher, check the guest editors, and check that the special issue appears on the journal’s official website and that the review process is clear.
A trustworthy call should not make those details difficult to find.
How should you read a CFP?
Open a long CFP and the first instinct is often to scan it for the submission date.
That is a good place to start, but not to stop.
Start with the dates
One CFP may give you a single deadline, while others give you four: abstract, full paper, decision, and final version.
Each date marks a distinct step.
NeurIPS adds another detail: Anywhere on Earth, or AoE. A deadline of 11:59 p.m. AoE does not mean 11:59 p.m. in your time zone. The deadline closes only after that date has ended in the last time zone on Earth.
That difference can amount to several hours.
So when you write the deadline in your calendar, do not copy the date alone. Convert the time too.
Find where your paper fits
Once the dates are clear, turn to the topics.
Suppose your paper studies energy-efficient client selection in hierarchical distributed learning. The conference offers tracks on distributed learning, edge computing, networking, optimization, and artificial intelligence.
All five sound related. They are not equally good choices.
Ask what problem the paper is really trying to solve.
If the study is about reducing resource use while edge devices take part in distributed learning, distributed learning or edge computing may describe it better than a broad artificial intelligence track.
This is not a keyword contest. A track does not become a good fit because its description repeats three words from your abstract.
The connection has to make sense.
Your title and abstract should make that connection visible. If communication costs and limited device resources are central to the paper, say so because they are central to the research, not because the CFP uses the same terms.
Fukuro can help here by comparing the abstract with the topics named in a CFP and showing where the connection is weak. It can tell you that a paper looks poorly matched to a track. It cannot tell you whether reviewers will accept the work.
That judgment comes later.
Check the formatting rules
Now imagine that you have found the right track. The paper fits. The research is ready.
There is still the submission itself.
Perhaps the conference allows eight pages, and your paper has nine. Perhaps it uses double-blind review, and your name is still on the first page. The supplementary file may exceed the maximum allowed size.
None of these problems concern the quality of the research. Any of them can still cause trouble before the review begins.
Under single-blind review, reviewers know who wrote the paper.
Under double-blind review, they do not. The authors remove names and affiliations and take care that references to earlier work do not reveal their identities.
Look beyond the manuscript
Some rules do not concern the paper file at all.
Imagine the conference accepts your paper, but it requires one author to register, travel to the event, and present it. Those conditions matter before submission, especially when travel is expensive or difficult.
The same is true for rules about submitting the work elsewhere.
A paper already under review at a journal may not be eligible for a conference that forbids simultaneous submission. In that situation, the problem is not the template or the page limit. The paper cannot enter both review processes simultaneously.
The small paragraphs near the end of a CFP can therefore matter as much as the deadline printed at the top.
What does it mean to respond to a CFP?
You rarely respond to a CFP by writing back.
You respond by submitting the work it asks for.
The paper belongs in the venue, follows its rules, and arrives on time. The title and abstract should make it easy for a reader to understand what the research is about and why it fits there.
That does not guarantee acceptance.
A reviewer can disagree with the methods or the contribution. Another paper could make a stronger case. A conference can receive more good work than it has room to publish.
Reading the CFP cannot change any of that.
What it can do is prevent another failure: a paper lost because the abstract was late, the track was wrong, the names were left in a double-blind submission, or a rule was missed.
The CFP tells you what kind of work the venue is asking for.
The rest is deciding whether your paper answers that call.
References and useful policies
- NeurIPS. Call for Papers, 2026 conference.
- AACR. Annual Meeting 2027.
- Retraction Watch. Wiley and Hindawi to retract 1,200 more papers for compromised peer review.
- ScienceDirect. Browse calls for papers for special issues.
- Elsevier. Publishing in a special issue.
- Think. Check. Attend. A checklist for identifying legitimate conferences.
- Think. Check. Submit. A checklist for choosing a trusted journal.
- H-Net Networks. Calls for Papers.