Talking to a Know-It-All GPT or a Second-Guesser Claude? How Repair reveals unreliable Multi-Turn Behavior in LLMs
For researchers and developers of conversational AI, this work highlights that LLMs exhibit distinct and unreliable repair behaviors in multi-turn interactions, which is a critical but underexplored aspect of human-LLM communication.
The study investigates how LLMs handle repair (trouble resolution) in multi-turn dialogues about math problems, finding that models vary from being resistant to repair to being easily manipulated, and that behavior becomes less predictable beyond single turns.
Repair, an important resource for resolving trouble in human-human conversation, remains underexplored in human-LLM interaction. In this study, we investigate how LLMs engage in the interactive process of repair in multi-turn dialogues around solvable and unsolvable math questions. We examine whether models initiate repair themselves and how they respond to user-initiated repair. Our results show strong differences across models: reactions range from being almost completely resistant to (appropriate) repair attempts to being highly susceptible and easily manipulated. We further demonstrate that once conversations extend beyond a single turn, model behavior becomes more distinctive and less predictable across systems. Overall, our findings indicate that each tested LLM exhibits its own characteristic form of unreliability in the context of repair.