Topics

This workshop will specifically focus on the opportunities and challenges of LLMs in real-world Machine Learning applications and will cover topics such as synthetic data generation, model evaluation and testing, fairness, debiasing, cold-start, distillation, and prompting. It will also welcome stories of LLM usage in specific applications, such as Chat, Search, Recommendations, Personalization, and/or Spoken Language Understanding. The goal is to bring together industry and academia researchers working on applied research problems in the LLM research space, fostering knowledge sharing and creating a discussion that paves the way to a societally beneficial exploitation of this technology. Possible topics are (but not limited to):

  • Synthetic data generation
  • LLM Testing and Evaluation
  • Fairness, accountability, trust
  • Green and Sustainable LLM
  • Debiasing
  • Cold-start
  • Knowledge Distillation
  • Customization / Personalization
  • Prompting
  • Explanation and Contextualization
  • LLMs applications (e.g. conversational AI, search, recommendations, natural language understanding, information extraction, summarization, Speech recognition systems)
  • Scalability
  • LLM using Human Feedback (RLHF)

Formats

We welcome original contributions describing applied research work in the LLM research field. Follow-up papers of more theoretical articles that describe specific challenges of productionizing LLMs in real-world applications are also welcome.

Contributions can be: full papers (up to 8 pages excluding references), or short papers (up to 4 pages excluding references). Papers must be written in English and formatted according to the Springer LNCS guidelines. Author instructions, style files, and the copyright form can be downloaded here, following the ECML-PKDD 2023 submission guidelines.

Papers must be written in English and formatted according to the Springer LNCS guidelines. Author instructions, style files, and the copyright form can be downloaded here, following the ECML-PKDD 2023 submission guidelines. The submissions will undergo a double-blind review process and be judged according to quality, relevance, scientific soundness and potential impact. Hence, authors should not disclose their names or affiliations and should do their best to avoid doing so indirectly in the text.

At least one author of each accepted paper must have a full registration and attend in-person to present the paper.

Electronic submissions will be handled via Microsoft CMT:

  1. login here
  2. click submission link
  3. click on “+ Create new submission…” button
  4. select “Challenges and Opportunities of Large Language Models in Real-World Machine Learning Applications”

Important Dates

Note: Submissions are to be done using the Microsoft CMT system (add link)

  • Jun 12th Jun 19th Jun 29th: Paper submission deadline
  • Jul 12th: Author notification
  • TBA Camera-ready deadline
  • Sep 18th: COLLM 2023 Workshop