CogniPair - Dynamic LLM Matching Algorithm in Chaotic Environments Mimicking Human Cognitive Processes for Relationship Pairing
Keywords: Large Language Models, Dating Algorithms, Human-like Reasoning, Context-aware Analysis, Simulating Characters, Machine Psychology
TL;DR: This paper introduces a framework using Large Language Models to simulate human characters and improve matchmaking by understanding nuanced personalities and social connections.
Abstract: Dating applications in the digital era have transformed how people connect, yet they often fall short in simulating the comprehensive character and fostering truly compatible relationships due to their reliance on quantitative data. This paper proposes a novel framework to simulate human characters by leveraging Large Language Models (LLMs) to enhance matchmaking by understanding the nuanced fabric of human personality and social connections. Traditional algorithms often lack the depth needed for personalized matchmaking, whereas LLMs offer sophisticated linguistic and cognitive capabilities to simulate a person and complicated personal decisions. Our framework introduces a multi-agent system comprising the Persona, Preference, and Dating Memory modules, allowing for dynamic and nuanced user interactions. This approach addresses the limitations of conventional LLM frameworks by capturing detailed personal attributes, updating preferences, and learning from past interactions. Our system enhances the relevance and effectiveness of match recommendations, focusing on emotional compatibility and shared values, providing a more personalized and responsive user experience in the dating domain.
Primary Area: generative models
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 11570
Loading