We invite proposals for offering evaluation tracks at FIRE 2023.
FIRE 2023 is the 15th edition of the annual meeting of Forum for Information Retrieval Evaluation ( fire.irsi.res.in). Since its inception in 2008, FIRE has had a strong focus on shared tasks similar to those offered at Evaluation forums like TREC, CLEF, and NTCIR. The shared tasks focus on solving specific problems in the area information access and, more importantly help in generating evaluation datasets for the research community.
It is not required for the tasks to focus on a specific language, and they can broadly cover any problem in the fields related to IR, NLP, multi-modal information access, and ML. However, the organizers especially encourage proposals for tracks related to South Asian, African, and middle Middle Eastern languages. In the past, FIRE has hosted tracks from Arabic, Persian, German, Russian, and Urdu languages besides several Indian languages. We aim to continue these efforts and include more language groups from these regions. For knowing more about tracks in past FIRE meetings, you can visit fire.irsi.res.in
Informal inquiries can also be sent to the track chairs.
Please include the following details in your proposal:
1. Track name
2. Track description
3. Broad categories of the track (eg: IR, NPL, ML, etc.). More than one categories can be mentioned.
4. Use case/s
5. Target Audience and number of expected submissions
6. Data(*) (fair details)
7. Evaluation plan
8. Timeline: Please try to align with the FIRE conference dates as given below
9. Organizer/s Details:
10. Prior experience in organizing shared task/workshop at relevant venues
|Track proposals due|
|Track acceptance notification due|
|25th May, 2023||Open track websites and training data release|
|10th July, 2023||Test data release|
|1st August, 2023||Run submission deadline|
|15th August, 2023||Results declared|
|15th September, 2023||Working notes due|
|15th October, 2023||Camera ready copies of working notes and overview paper due|
Note: Tentatively in December 2023.
Please send these details in a pdf format to firstname.lastname@example.org with a copy to email@example.com and firstname.lastname@example.org
(*) We require that after FIRE, the data be made publicly available through Information Retrieval Society of India. In case of data that can not be distributed publicly (e.g., Twitter data), a unique identifier that can be used to recreate the original corpus can be provided (e.g., tweet ids in case of Twitter data). This disbursal will be governed by a copyright form, which the users have to sign before getting the data. A sample form is available at ( http://fire.irsi.res.in/fire/static/data ).
In case it is not preferable/possible for the track organizers to share the data, please mention this in the proposal with specific concerns. Exceptions can be made for tracks where data from industry is used or in case of other serious legal or ethical concerns.
The aim of organizing these tracks at FIRE is to have debates and discussions on focused topics and give feedback to participants. As a result, at least one of the track organizers from each track is expected to attend FIRE and present the overview of track in person. In case of non-attendance of any of the organizers, the team will not be allowed to offer a track next year. This year, depending on the covid-19 situation, it is likely that a remote presentation will be allowed.
We will try to provide student volunteers for support in the proposed tracks. They will basically be undergraduate students interested in IR and related fields and can help with corpus creation, evaluation, correspondence with participants, etc. If you require any such support, kindly mention that in the track proposals along with the number of students required.
Hoping to have an enthusiastic response from your end.
Thomas Mandl (Universitat Hildesheim, Germany)
Kripabandhu Ghosh (IISER Kolkata, India)