InterIntent: Investigating Social Intelligence of LLMs via Intention Understanding in an Interactive Game Context
This work addresses the evaluation gap in social intelligence for LLM researchers, though it is incremental as it builds on existing game-based tests.
The paper tackled the problem of evaluating social intelligence in large language models (LLMs) by developing the InterIntent framework to assess intention understanding in an interactive game context, finding that LLMs achieved 88% accuracy in intention selection but trailed human performance by 20% in intention inference.
Large language models (LLMs) have demonstrated the potential to mimic human social intelligence. However, most studies focus on simplistic and static self-report or performance-based tests, which limits the depth and validity of the analysis. In this paper, we developed a novel framework, InterIntent, to assess LLMs' social intelligence by mapping their ability to understand and manage intentions in a game setting. We focus on four dimensions of social intelligence: situational awareness, self-regulation, self-awareness, and theory of mind. Each dimension is linked to a specific game task: intention selection, intention following, intention summarization, and intention guessing. Our findings indicate that while LLMs exhibit high proficiency in selecting intentions, achieving an accuracy of 88%, their ability to infer the intentions of others is significantly weaker, trailing human performance by 20%. Additionally, game performance correlates with intention understanding, highlighting the importance of the four components towards success in this game. These findings underline the crucial role of intention understanding in evaluating LLMs' social intelligence and highlight the potential of using social deduction games as a complex testbed to enhance LLM evaluation. InterIntent contributes a structured approach to bridging the evaluation gap in social intelligence within multiplayer games.