NLP4HR ワークショップ
Overview
NLP4HRワークショップは、自然言語処理（NLP）と人事（HR）の交差点で最先端の技術と実践が出会う場です。本ワークショップでは、研究者、業界リーダー、実務者が一堂に会し、採用、バイアスの軽減、パフォーマンス管理、AIを活用したキャリア支援など、HRにおける重要な課題に取り組みます。基調講演、インタラクティブなディスカッション、最先端の研究発表を通じて、NLPがHRの未来をどのように変革できるかを探ります。また、公平性、プライバシー、透明性といった倫理的課題にも焦点を当て、AIのより良い活用方法を模索します。
Program
8:30am – 8:40: Opening
8:40am – 9:30am: Invited Talk: David Graus – Recommender systems, bias, and bias mitigation in hiring
Abstract: In my talk, I will explore the intersection of recommender systems, bias, and bias mitigation in hiring, focusing on Natural Language Processing (NLP) challenges and solutions. I will do so by discussing recommender systems for matching jobs and job seekers at Randstad. My talk will cover aspects of incorporating textual features in recommender systems, the challenges of bias from textual features, and several strategies for mitigating such bias, drawing from our experiences, literature and our own research.
9:30am – 10:30am: Oral Presentations 1
9:30am – 9:40am: HR-MultiWOZ: A Task Oriented Dialogue (TOD) Dataset for HR LLM Agent
- Weijie Xu, Zicheng Huang, Wenxiang Hu, Xi Fang, Rajesh Kumar Cherukuri, Naumaan Nayyar, Lorenzo Malandri, Srinivasan H. Sengamedu
- Antoine Magron, Anna Dai, Mike Zhang, Syrielle Montariol, Antoine Bosselut
- Charlie Campanella, Rob van der Goot
- Khanh Cao Nguyen, Mike Zhang, Syrielle Montariol, Antoine Bosselut
10:30am – 11:00am: Break
11:00am – 11:50am: Invited Talk by Barbara Plank – Natural Language Processing for Human Resources: Why, How and What’s Next
Abstract: Natural Language Processing (NLP) can offer valuable insights and technology for Human Resource (HR) practices by streamlining tasks such as job posting analysis, candidate matching, and skill assessment. In this talk, I will delve into the intersection of NLP and HR, exploring the motivations behind leveraging NLP techniques for job market analysis, providing an overview of the methodologies employed and datasets available today, as well as challenges and future directions for this evolving field of computational job market analysis. This includes showcasing some of our own work from the MultiSkill project aimed at multilingual information extraction from job postings.
11:50am – 12:20pm: Oral Presentations 2
- Elena Senger, Mike Zhang, Rob van der Goot, Barbara Plank
12:05pm -12:20pm: Aspect-Based Sentiment Analysis for Open-Ended HR Survey Responses
- Lois Rink, Job Meijdam, David Graus
12:20pm – 1:00pm Lunch Break
1:00pm – 1:50pm: Invited Talk: Marko Grobelnik
1:50pm – 3:05pm: Panel Discussion by Madhu Kurup, Malihe Alikhani, Marko Grobelnik, and Navid Rekab-Saz
3:05pm – 4:00pm: Poster Presentations
- Aspect-Based Sentiment Analysis for Open-Ended HR Survey Responses. Lois Rink, Job Meijdam, David Graus
- Deep Learning-based Computational Job Market Analysis: A Survey on Skill Extraction and Classification from Job Postings . Elena Senger, Mike Zhang, Rob van der Goot, Barbara Plank
- Entity Linking in the Job Market Domain. Mike Zhang, Rob van der Goot, Barbara Plank
- NNOSE: Nearest Neighbor Occupational Skill Extraction. Mike Zhang, Rob van der Goot, Min-Yen Kan, Barbara Plank
4:00pm – 4:10pm: Closing
Review Committee
Aisha Aijaz (Indraprastha Institute of Information Technology, Delhi)
Benjamin Clavié (Bright Network)
Bo Kang (Ghent University)
Hiroshi Matsuda (Megagon Labs)
Kaixin Ma (Tencent AI Lab)
Lorenzo Malandri (University of Milan – Bicocca)
Mesut Kaya (Aalborg University Copenhagen)
Michiharu Yamashita (Pennsylvania State University)
Mike Zhang (IT University of Copenhagen)
Nidhi Goyal (Indraprastha Institute of Information Technology, Delhi)
Prakhar Gupta (Google)
Sairam Gurajada (Megagon Labs)
Sajjadur Rahman (Megagon Labs)
Seiji Maekawa (Megagon Labs)
Takashi Kodama (Kyoto University)
Takuya Makino (Megagon Labs)
Weijie Xu (Amazon)
Yoshinari Fujinuma (AWS AI Labs)
Panelists
Madhu Kurup
Malihe Alikhani
Marko Grobelnik - Distinguished Speaker
Navid Rekab-saz
Accepted Papers
Big City Bias: Evaluating the Impact of Metropolitan Size on Computational Job Market Abilities of Language Models by Charlie Campanella, Rob van der Goot
JobSkape: A Framework for Generating Synthetic Job Postings to Enhance Skill Matching by Antoine Magron, Anna Dai, Mike Zhang, Syrielle Montariol, Antoine Bosselut
Rethinking Skill Extraction in the Job Market Domain using Large Language Models by Khanh Cao Nguyen, Mike Zhang, Syrielle Montariol, Antoine Bosselut
Aspect-Based Sentiment Analysis for Open-Ended HR Survey Responses by Lois Rink, Job Meijdam, David Graus
Deep Learning-based Computational Job Market Analysis: A Survey on Skill Extraction and Classification from Job Postings by Elena Senger, Mike Zhang, Rob van der Goot, Barbara Plank
HR-MultiWOZ: A Task Oriented Dialogue (TOD) Dataset for HR LLM Agent by Weijie Xu, Zicheng Huang, Wenxiang Hu, Xi Fang, Rajesh Kumar Cherukuri, Naumaan Nayyar, Lorenzo Malandri, Srinivasan H. Sengamedu
Mentorship Program
We will be hosting a mentorship program to facilitate exchange between potential workshop participants and experts working in areas relevant to the workshop. The goal of the mentorship program is to support underrepresented communities of NLP, as well as junior researchers. Mentors are expected to guide mentees as they prepare submissions for the workshop. Mentees are expected to initiate contact with their assigned mentor and prepare the necessary material for discussion and submission. Mentorship lasts the duration of the program, but the mentor-mentee can of course continue to meet as frequently as they commonly agree upon. We recommend mentees set up a meeting within the first week after the mentor-mentee match is made, discuss their mutual expectations and plan their subsequent interactions. Some example interactions are: 1) iterate on the core idea of the paper, ensuring the problem is relevant and interesting to the community, proposed techniques are well-suited, and experimental results are comprehensive. 2) Discuss in detail related work to ensure submitted work is novel and relevant. 3) Identify areas to expand the paper: discuss short-term open research problems relevant to the submission, and 4) receive feedback on the manuscript to improve the quality of the submission.
Applications to the mentorship program are due December 15th, 2023.
Applications and more details can be found here.
Scholarship Program
The workshop will cover the registration costs (full registration EACL + Workshops + Tutorials) for up to 4 students interested in our workshop. If more than 4 students apply, the organizing committee will prioritize all traditionally underrepresented groups in technology, and if more than 4 authors from underrepresented groups apply, priority will be given to authors of papers accepted/submitted to the workshop. If any further selection is still needed, it will be based on a first-come, first-served basis.
Call For Paper
We invite submissions of both long and short papers that present original and previously unpublished research addressing the challenges associated with the application of NLP for HR. We also consider non-archival submissions upon request (*Please inform the organizers before submitting). Accepted papers will be presented either through oral presentations or poster sessions and will be included in the EACL proceedings as workshop papers.
All regular papers and short papers should adhere to the EACL 2024 submission guidelines, with the only exception being the mandatory inclusion of a limitations section. While we strongly encourage authors to discuss the limitations of their work, we will not reject papers solely for lacking a limitations section.
- Long papers can have up to 8 pages of content, with unlimited references and appendices.
- Accepted long papers will also receive an additional page of content (up to 9 pages) in the proceedings to incorporate reviewers’ comments.
- Short papers can have up to 4 pages of content, with unlimited references and appendices.
- Accepted short papers will also receive an additional page of content (up to 5 pages) in the proceedings to incorporate reviewers’ comments.
- All submissions must follow the official ACL style format, which can be found here.
- Unlike the main track of EACL 2024, papers without a limitations section will still undergo review for the NLP4HR workshop. Nevertheless, we strongly encourage authors to address the limitations of their work in a dedicated section titled “Limitations.” This section should be placed at the end of the main content of the paper, following the discussion/conclusions section and preceding the references, and will not count towards the page limit.
For more detailed information, please refer to the EACL 2024 website.
Topics of Interest:
This workshop is centered around various aspects of applying NLP in the HR domain, including but not limited to:
- Automated creation of HR knowledge bases.
- Parsing, extracting, or inferring information from HR documents (such as job descriptions, resumes, and CVs).
- Techniques for handling semi-structured HR documents.
- Learning representations for HR entities (e.g., jobs, job seekers, employers).
- Search and recommendation systems tailored to HR.
- Dialogue-based HR assistants, agents, or systems.
- Natural language generation for HR purposes.
- Question answering for HR-related queries.
- Detecting and addressing bias in HR.
Important Notes:
- No dual submission: NLP4HR will not consider any paper that is currently under review in a journal, another conference, or workshop at the time of submission, and submitted papers must not be simultaneously submitted elsewhere during the review period.
- Presentation: All accepted papers must be presented either in person or virtually at the workshop. At least one author of each accepted paper must register for EACL 2024 and participate in the workshop.
