Go to ACL ARR 2025 May homepageTurnBench-MS: A Benchmark for Evaluating Multi-Turn, Multi-Step Reasoning in Large Language Models
Abstract: Despite impressive advances in large language models (LLMs), existing benchmarks often focus on single-turn or single-step tasks, failing to capture the kind of iterative reasoning required in real-world settings. To address this limitation, we introduce \textbf{TurnBench}, a novel benchmark that evaluates multi-turn, multi-step reasoning through an interactive code-breaking task inspired by a ``Turing Machine Board Game.'' In each episode, a model must uncover hidden logical or arithmetic rules by making sequential guesses, receiving structured feedback, and integrating clues across multiple rounds. This dynamic setup requires models to reason over time, adapt based on past information, and maintain consistency across steps—capabilities underexplored in current benchmarks. TurnBench includes two modes: \textit{Classic}, which tests standard reasoning, and \textit{Nightmare}, which introduces increased complexity and requires robust inferential chains. To support fine-grained analysis, we provide ground-truth annotations for intermediate reasoning steps. Our evaluation of state-of-the-art LLMs reveals significant gaps: GPT-4-mini achieves 81.5\% accuracy in Classic mode, but performance drops to 17.8\% in Nightmare mode. In contrast, human participants achieve 100\% in both, underscoring the challenge TurnBench poses to current models. By incorporating feedback loops and hiding task rules, TurnBench reduces contamination risks and provides a rigorous testbed for diagnosing and advancing multi-step, multi-turn reasoning in LLMs.
Paper Type: Long
Research Area: Resources and Evaluation
Research Area Keywords: Reasoning, LLM, Evaluation,
Contribution Types: Data resources, Data analysis
Languages Studied: English
Submission Number: 7381
Loading