General information on ACL Anthology assignments for event (proceedings) chairs: (Updated as of 13 Oct 2015)

File naming

Currently, we assign each paper in the ACL Anthology a unique 8-character identifier, so that we can easily locate papers and reference them. The identifier is assigned by the current ACL Anthology Editor and is in the form of a letter, two digit year and a 4 digit paper ID (e.g., W11-3801). For large or prominent events related to ACL or its sister organizations that recur on a yearly basis, we use a separate lettered prefix, where all 4 digits of the paper ID can be used (usually accommodating separate volumes or issue numbers in the event or journal, each capable of referencing up to 999 papers).

For all other smaller events, we use the 'W' prefix. These events accommodate up to 99 papers. If there are multiple volumes for a such smaller events, they are assigned as two separate volumes (e.g., shared tasks, student papers).

The file names fit within the 8.3 DOS naming constraint for maximum portability, as recommended by Adobe. PDF filenames are globally unique, to support subsetting (saving an ad hoc collection of papers to a single directory). The numeric fields within filenames are zero-padded to ensure filenames have fixed width.

CodesSetsFilename ExampleComments
A, C, D, E, H, I, L, M, N, P, S, T, XProceedings
(syy-xnnn)
P90-1001This is the first paper appearing in the first (or only) volume of the P90 proceedings; most proceedings have just one volume; in rare cases a proceedings volume has a supplement which should be numbered as a separate volume; in rare cases multiple proceedings volumes are bound into one (N00/A00) and these should be treated as separate volumes. Each conference proceedings may have up to 999 papers; conferences with more papers than this upper limit should consult the ACL Anthology Editor on how to split the proceedings into separate volumes. Proceedings chairs of conferences may choose at their discretion how they would like to partition volumes, although segmentations involving main/full papers, short papers, demonstrations and tutorial abstracts are most common.
J,F,QJournal
(jyy-xnnn)
J90-2001This is for the first paper in the second issue of J90; For combined issues, like 3/4, use the first number of the sequence. (E.g. if a journal year consists of combined issues 1/2 and 3/4, use J90-1 and J90-3 only). The 'F' prefix is for the Finite String, a newsletter (now obsolete) that used to be part of the Journal.
WWorkshops and Smaller Events
(wyy-xxnn)
W90-0201This is for the first paper in the second workshop in 1990; there is space for up to 100 workshops per year, and up to 99 papers per workshop. If a workshop exceeds over 99 papers in a year, please consult the ACL Anthology Editor (currently Min-Yen Kan) first. In this case, a separate set letter code may be established for the venue.

Workshop chairs should contact the ACL Anthology Editor to receive their workshop number offset (the 'xx' portion of the ACL Anthology ID). If your workshop is attached to a conference as a satellite event, please contact the proceedings chair for the main conference to receive the offset ID, as it is easiest to allocate offsets as a whole block. Conversely, if you are the proceedings chair for a conference that has satellite workshops, please contact the ACL Anthology Editor with the final list of titles of the workshops (make certain the workshops will actually be run) so that the editor can allocate a suitable block of offsets to the workshops.

Paper numbers

A key notion in file naming is the paper number. The general rule is that papers will be numbered consecutively within the bound volume in which they appear. When a proceeding is divided into multiple volumes, paper number will begin from number '1', with each new volume. When multiple proceedings are bound into a single volume (e.g. N00), they will be treated as multiple volumes.

Any front matter is given the number '0'. Any back matter is given the number one more than the last paper in the volume. Front and back matter that appears internally to a volume (e.g. in N00) will be counted just like an ordinary paper.

Metadata

Anthology index pages are generated from an XML file containing information about volumes, paper titles and authors. Each publication has an XML file, stored alongside the scanned images. Proceedings chairs using the aclpub publication package have a build target that will build a richer version of the metadata below. Here is a fragment of P/P03/P03.xml, which satisfies these minimal requirements, generated automatically from the conference CD-ROM:

<?xml version="1.0" encoding="UTF-8" ?>
 <volume id="P03">
   <paper id="1000">
        <title>Proceedings of the 41st Annual Meeting of the Association for Computational Linguistics</title>
   </paper>

   <paper id="1001">
        <title>Offline Strategies for Online Question Answering: Answering Questions Before They Are Asked</title>
        <author>Michael Fleischman</author>
        <author>Eduard Hovy</author>
        <author>Abdessamad Echihabi</author>
   </paper>

   <paper id="1002">
        <title>Using Predicate-Argument Structures for Information Extraction</title>
        <author>Mihai Surdeanu</author>
        <author>Sanda Harabagiu</author>
        <author>John Williams</author>
        <author>Paul Aarseth</author>
   </paper>

   <paper id="1003">
        <title>A Noisy-Channel Approach to Question Answering</title>
        <author>Abdessamad Echihabi</author>
        <author>Daniel Marcu</author>
   </paper>

   ...

 </volume>

A more comprehensive example can be found at:

http://www.aclweb.org/anthology/P/P15/P15.xml
(metadata for ACL 2015)

These metadata files are publicly visible (but not linked) from the current Anthology.

Other details:

  1. URLs for the Anthology (for START as well as the Anthology XML) look like:

    http://www.aclweb.org/anthology/W11-38xx

    and not

    http://www.aclweb.org/anthology/W/W11/W11-38xx.pdf

    (note, no .pdf and no intermediate path)

    where xx is the paper ID. 00 is used for the frontmatter, 01 for the first paper, etc., and any author index is listed as the final paper. In START you may see (and need to provide) this coded using C-like format strings:

    http://www.aclweb.org/anthology/W11-38%02d
  2. The data in the ACL Anthology is (almost) completely derived from the XML form of the paper and volume metadata that is used to generate it. If you can create this XML representation for your event, it will be sufficient information for us to generate the Anthology representation (Web pages) for your event. For events using START version 2 (via softconf.com), which is the default publication / event manager software for ACL, this will be easy, as we have transformation scripts to ingest directly from the proceedings.tgz file that can be one-click generated from START. For other events, you will need to provide this data on your own (see the Metadata section above).
  3. For copyright transfers, you can use the default forms at:
    http://wing.comp.nus.edu.sg/~antho/acl-copyright-form-rev2011.pdf
    http://wing.comp.nus.edu.sg/~antho/acl-copyright-form-rev2011.doc

    Ask authors to sign the form, scan a B/W copy at 150 to 200 DPI for you as a .pdf (or .jpg or .png if otherwise not possible). Name the forms using the ACL Anthology identifiers, and send me a separate .zip or .tgz archive as you would for the proceedings of your event (e.g., a file in the archive might be copyright-transfers/P11-1001. pdf).

    Note that if you are using the START system (softconf.com), there is a digital form of this copyright transfer form, so you may not have to ask the authors to print, sign and submit to you. You may be able to simply download the .zip or .tgz version of the archived forms and submit them to me. Please note this when submitting.

    For both current and legacy events, it is good practice for the organizers to attempt to obtain copyright transfers for their materials, but we will ingest materials even if no copyright transfers are on file.

  4. If you need to assign ISBN numbers, please provide the exact titles of each volume to be assigned an ISBN and forward this information to Priscilla Rasmussen, ACL Business Manager.
  5. If there are supplemental attachments, software or datasets, also please let me know about them when they are finalized so that we can have them published in the Anthology in a timely manner. You'll need to provide a zipfile or other filetype (of a three letter extension only) of each supplement, where the files are named by their paper's Anthology ID (e.g., W11-1001.Dataset.zip)

    Current types supported by the Anthology are:

    • Attachment (this is for generic attachments; best to be specific and avoid using this when possible)

    • Dataset
    • Note
    • Poster
    • Presentation
    • Software

    Videos are also accepted but only as valid hyperlinks.

    Any attachment is limited to 30 MB or a valid hyperlink (which should be maintained in perpetuity, but reasonably maintained for at least 5 years). Hyperlinks may lead either to the actual attachment or to a valid containing "home page" for the resource.

    We prefer actual attachments over hyperlinks as we can serve them from the Anthology directly, but understand that some materials cannot fit within this size. Attachments do not need to be author-anonymized (although during the review process, program chairs may ask for this -- authors should try to include provenance information with their attachments).

  6. If you get requests from authors needing to post errata or revised versions of the papers, or supplemental attachments after the publication of the proceedings, please ask them to submit revisions to the current ACL Anthology editor directly (see Corrections).

    If you wish to incorporate edits from your pipeline, we will also accept revisions from proceedings chairs of workshops (to replace the original version, or to create a new version) if the ACL footer (i.e. "Proceedings of ..." on header page) appears in the PDF with the correct page numbers, only before the publication date. After the publication date, the full volume cannot be replaced or revised.

  7. By default, the publication of papers associated with a current event in the Anthology will be on the first day of the event (inclusive of workshops or tutorials). If you prefer to have it published on a different date, please inform us.
  8. Please list the ACL SIGs endorsing or sponsoring each event, and for each event, where not completely clear, give a short, acronym form of the event name, for indexing (e.g., "ACL", "NAACL", "SocialNLP"). Acronym forms should be recognizable and consistent through consecutive (multi-)annual events. Please check the beta Anthology (http://aclanthology.info/catalog?utf8=%E2%9C%93&search_field=all_fields&q=# , see venue facet) to check whether the venue acronym has been used before). Please provide this information.
  9. For ingestion from events managed by Easy Chair, Nils Blomqvist has created a easy2acl script, which should make the transformation to the aclpub format easier.
  10. Q: What data from START (softconf.com) does the Anthology ingestion process actually use?
    A: The Anthology software uses a transformation script to revise the distributed .bib files into the appropriate XML metadata used by the Anthology. We typically run this on the full volume .bib file, provided at the cdrom/ directory that is adistributed by START. We primarily use three sources of information that is just renamed from those sources:
    • the bib files under cdrom/bib
    • the pdf files under cdrom/pdf
    • the whole volume bib/pdf files under cdrom

    We also use the cdrom/attachments when they are provided (optional)

  11. Other general information on ACL Anthology assignment is included on the following page (although outdated). Please read it. http://www.aclweb.org/anthology/importing.html