Call for Participation TalentCLEF Workshop/Lab (CLEF 2026)
Skill and Job Title Intelligence for Human Capital Management
https://talentclef.github.io/talentclef/
TalentCLEF is an initiative to advance Natural Language Processing (NLP) in
Human Capital Management (HCM). It aims to create a public benchmark for
model evaluation and promote collaboration to develop fair, multilingual,
and flexible systems that improve Human Resources (HR) practices across
different industries.
📅 Registration Deadline: April 23, 2026
Key information:
-
Web: https://talentclef.github.io/talentclef/
-
Data: https://doi.org/10.5281/zenodo.17625261
-
Registration: https://clef2026.clef-initiative.eu/
Why TalentCLEF?
The labor market is undergoing a profound transformation. Roles evolve
faster than ever, skills become obsolete in years rather than decades, and
organizations operate in increasingly multilingual and global environments.
At the same time, Artificial Intelligence, and in particular Natural
Language Processing and Large Language Models, is reshaping how talent is
identified, matched, and developed, enabling organizations to analyze job
requirements and candidate profiles at scale and to respond more rapidly to
evolving workforce needs.
Despite this progress, talent intelligence systems still face major
challenges. Much of the recent progress in NLP for Human Capital Management
relies on private or restricted datasets, together with a lack of
standardized evaluation frameworks, which hinders reproducibility and
prevents the establishment of robust benchmarks. In addition, many
approaches struggle to operate reliably in multilingual and cross-lingual
settings, and growing concerns remain regarding fairness and ethical
robustness in automated decision-making systems.
TalentCLEF 2026 addresses these challenges by providing open and
privacy-preserving benchmarks, realistic use cases grounded in Human
Capital Management workflows, and a shared evaluation framework that
enables meaningful comparison of methods. By bridging research innovation
with real-world requirements, TalentCLEF aims to foster the development of
NLP systems that are not only accurate, but also transparent, fair, and
applicable in practice.
Tasks Overview
TalentCLEF 2026 is structured into two independent but complementary tasks,
allowing participants to compete in one or both.
-
Task A - Contextualized Job-Person Matching: Task A challenges
participants to develop systems capable of identifying and ranking the most
suitable candidate résumés for a given job offer, based on their overall
relevance to the position. Unlike approaches that focus on isolated
entities such as job titles or individual skills, this task emphasizes
context-aware matching between full job descriptions and candidate
profiles. The task is framed as an information retrieval problem, where
each job offer is matched against a fixed set of candidate résumés, which
are returned as a ranked list sorted by relevance. The evaluation is
conducted in English and Spanish, with an additional cross-lingual setting,
reflecting realistic recruitment scenarios in multilingual environments and
encouraging the development of robust, language-aware matching models.
-
Task B - Job-Skill Matching with Skill Type Classification: Task B
focuses on automatically identifying the skills associated with a given job
title and classifying each retrieved skill according to its role within the
job profile. Participants are required to retrieve relevant skills from a
predefined gazetteer, but also to determine whether each relevant skill is
core (required) or complementary (optional). This task reflects the growing
importance of skill-based representations in Human Capital Management,
supporting applications such as recruitment, workforce planning, and
upskilling. By combining skill retrieval with relevance-based
classification, Task B encourages approaches that go beyond simple matching
and capture the nuanced importance of different skills across job roles.
Schedule
-
2nd February 2026 - Development data available for Tasks A
-
2nd February 2026 - Training data available for Task B
-
16th February 2026 - Development data available for Task B
-
2nd March 2026 - Codabench Release for Task A and Task B
-
13th April - 3rd May 2026 - Evaluation Period for Task A and Task B
-
23rd April - Official registration deadline
-
7th May 2025 – Publication of Official Results
-
30th May 2025 – Submission of CLEF 2025 Participant Working Notes
(CEUR-WS)
-
27th June 2025 - Notification of Acceptance for Participant Papers
Publications and CLEF 2026 workshop
Teams participating in TalentCLEF will be invited to submit a system
description paper for the CLEF 2026 Working Notes proceedings, published on
CEUR-WS. Additionally, they will have the opportunity to present a brief
overview of their approach at the TalentCLEF 2026 workshop, which will be
co-located at CLEF 2026 and will take place in Jena, Germany, from
September 21st to 24th, 2026.
Main Organizers
-
Luis Gascó, Avature
-
Hermenegildo Fabregat, Avature
-
Laura García-Sardiña, Avature
-
Paula Estrella, Avature, Spain
-
Casimiro Pío Carrino, Avature, Spain
-
Jens-Joris Decorte, TechWolf
-
Matthias De Lange, TechWolf
-
Daniel Deniz Cerpa, Avature, Spain
-
Álvaro Rodrigo, Universidad Nacional de Educación a Distancia (UNED)
-
Rabih Zbib, Avature, Spain
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************
Posts
