MIRROR@IberLEF2026 Challenge: Generative AI for Motivational Interviewing

Deadline: May 11th

Participate in the MIRROR@IberLEF2026 challenge to develop Generative AI methods for improving behavior code classification in motivational interviews.

The MIRROR@IberLEF2026 challenge is now live, inviting participants to develop Generative AI methods for creating synthetic conversation turns to improve the performance of models trained to recognize behavior codes in motivational interviews.

The task involves producing high-quality datasets representing diverse clinical conversations to enhance the performance of a frozen baseline model used for behavior code classification.

• Participants are expected to provide three datasets of at most 100 labeled conversation turns.
• The fine-tuned models will be used to make predictions for a hold-out dataset.
• The considered pairs of behavior codes are: Simple reflection vs. Complex reflection, Open question vs. Closed question, and Persuasion vs. Giving Information.

Registration and submission details are available on the MIRROR challenge platform.

Key dates include:

• Mar 9th: Start of development phase
• May 1st: Start of final phase
• May 11th: Deadline for submission of runs
• Jun 8th: Deadline for paper submission

Tags: MIRROR@IberLEF2026, Generative AI, Motivational Interviewing, Behavior Code Classification, Synthetic Conversation Turns, AI Challenge, Natural Language Processing