TurnBench-MS: A Benchmark for Evaluating Multi-Turn, Multi-Step Reasoning in Large Language Models

• URL: http://arxiv.org/abs/2506.01341v1
• Date: Mon, 02 Jun 2025 05:47:50 GMT
• Title: TurnBench-MS: A Benchmark for Evaluating Multi-Turn, Multi-Step Reasoning in Large Language Models
• Authors: Yiran Zhang, Mo Wang, Xiaoyang Li, Kaixuan Ren, Chencheng Zhu, Usman Naseem,
• Abstract summary: We introduce TurnBench, a novel benchmark that evaluates multi-turn, multi-step reasoning through an interactive code-breaking task.<n>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.<n>TurnBench includes two modes: Classic, which tests standard reasoning, and Nightmare, which introduces increased complexity and requires robust inferential chains.
• Score: 5.6525926183880255
• License: http://creativecommons.org/licenses/by/4.0/
• 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 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: Classic, which tests standard reasoning, and 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: the best model 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.

Related papers

• Multimodal Fact-Level Attribution for Verifiable Reasoning [80.60864342985748]
  Multimodal large language models (MLLMs) are increasingly used for real-world tasks involving multi-step reasoning and long-form generation.<n>Existing multimodal grounding benchmarks and evaluation methods fail to assess attribution in complex multimodal reasoning.<n>We introduce MuRGAt, a benchmark for evaluating fact-level multimodal attribution in settings that require reasoning beyond direct observation.
  arXiv  Detail & Related papers  (2026-02-12T03:10:02Z)
• TSRBench: A Comprehensive Multi-task Multi-modal Time Series Reasoning Benchmark for Generalist Models [52.21298691935726]
  The ability to reason over time series is a fundamental skill for generalist models to solve practical problems.<n>To bridge this gap, we introduce TSRBench, a comprehensive benchmark designed to stress-test the full spectrum of time series reasoning capabilities.
  arXiv  Detail & Related papers  (2026-01-26T18:04:54Z)
• ArenaBencher: Automatic Benchmark Evolution via Multi-Model Competitive Evaluation [33.22383550511664]
  ArenaBencher is a model-agnostic framework for automatic benchmark evolution.<n>We apply ArenaBencher to math problem solving, commonsense reasoning, and safety domains.
  arXiv  Detail & Related papers  (2025-10-09T17:59:55Z)
• MR$^2$-Bench: Going Beyond Matching to Reasoning in Multimodal Retrieval [86.35779264575154]
  Multimodal retrieval is becoming a crucial component of modern AI applications, yet its evaluation lags behind the demands of more realistic and challenging scenarios.<n>We introduce MR$2$-Bench, a reasoning-intensive benchmark for multimodal retrieval.
  arXiv  Detail & Related papers  (2025-09-30T15:09:14Z)
• seqBench: A Tunable Benchmark to Quantify Sequential Reasoning Limits of LLMs [1.0519693622157462]
  We introduce seqBench, a benchmark for probing sequential reasoning limits in Large Language Models (LLMs)<n>We find that even top-performing models systematically fail on seqBench's structured reasoning tasks despite minimal search complexity.
  arXiv  Detail & Related papers  (2025-09-21T01:32:13Z)
• STEPWISE-CODEX-Bench: Evaluating Complex Multi-Function Comprehension and Fine-Grained Execution Reasoning [6.282781900938977]
  We present STEPWISE-CODEX-Bench (SX-Bench), a novel benchmark for complex multi-function understanding and fine-grained execution reasoning.<n>SX-Bench is highly discriminative, even the state-of-the-art OpenAI-O3 achieves only 78.37 percent accuracy on Hard-Reasoning tasks.
  arXiv  Detail & Related papers  (2025-08-07T09:28:43Z)
• OST-Bench: Evaluating the Capabilities of MLLMs in Online Spatio-temporal Scene Understanding [50.72259772580637]
  We introduce OST-Bench, a benchmark designed to evaluate Online Spatio-Temporal understanding from the perspective of an agent.<n>Built on an efficient data collection pipeline, OST-Bench consists of 1.4k scenes and 10k question-answer pairs collected from ScanNet, Matterport3D, and ARKitScenes.<n>We find that both complex-based spatial reasoning demands and long-term memory retrieval requirements significantly drop model performance along two separate axes.
  arXiv  Detail & Related papers  (2025-07-10T17:56:07Z)
• Reasoning Multimodal Large Language Model: Data Contamination and Dynamic Evaluation [9.434966074326056]
  Multimodal Large Language Models (MLLMs) show impressive vision-language benchmark performance, yet growing concerns about data contamination risk masking true generalization.<n>We propose a novel dynamic evaluation framework to rigorously assess MLLM generalization, moving beyond static benchmarks.<n>We demonstrate that fine-tuning on simulated test data (extreme contamination) drastically sharpens task-specific performance but harms overall generalization.
  arXiv  Detail & Related papers  (2025-06-08T15:52:38Z)
• MORSE-500: A Programmatically Controllable Video Benchmark to Stress-Test Multimodal Reasoning [54.47710436807661]
  MORSE-500 is a video benchmark composed of 500 fully scripted clips embedded questions spanning six complementary reasoning categories.<n>Each instance is generated using deterministic Python scripts (Manim, Matplotlib, MoviePy), generative video models, and real footage.<n>Unlike static benchmarks that become obsolete once saturated, MORSE-500 is built to evolve.
  arXiv  Detail & Related papers  (2025-06-05T19:12:45Z)
• MTR-Bench: A Comprehensive Benchmark for Multi-Turn Reasoning Evaluation [56.87891213797931]
  We present MTR-Bench for Large Language Models' Multi-Turn Reasoning evaluation.<n>Comprising 4 classes, 40 tasks, and 3600 instances, MTR-Bench covers diverse reasoning capabilities.<n>MTR-Bench features fully-automated framework spanning both dataset constructions and model evaluations.
  arXiv  Detail & Related papers  (2025-05-21T17:59:12Z)
• Model Utility Law: Evaluating LLMs beyond Performance through Mechanism Interpretable Metric [99.56567010306807]
  Large Language Models (LLMs) have become indispensable across academia, industry, and daily applications.<n>One core challenge of evaluation in the large language model (LLM) era is the generalization issue.<n>We propose Model Utilization Index (MUI), a mechanism interpretability enhanced metric that complements traditional performance scores.
  arXiv  Detail & Related papers  (2025-04-10T04:09:47Z)
• FINEREASON: Evaluating and Improving LLMs' Deliberate Reasoning through Reflective Puzzle Solving [90.88021670297664]
  FINEREASON is a logic-puzzle benchmark for evaluation of large language models' reasoning capabilities.<n>We introduce two tasks: state checking, and state transition, for a comprehensive evaluation of how models assess the current situation and plan the next move.<n>We show that models trained on our state checking and transition data demonstrate gains in math reasoning by up to 5.1% on GSM8K.
  arXiv  Detail & Related papers  (2025-02-27T16:23:25Z)
• LR$^2$Bench: Evaluating Long-chain Reflective Reasoning Capabilities of Large Language Models via Constraint Satisfaction Problems [7.379503137362718]
  We introduce LR$2$Bench, a novel benchmark designed to evaluate the Long-chain Reflective Reasoning capabilities of Large Language Models (LLMs)<n>Our experimental results reveal that even the most advanced reasoning-specific models, such as DeepSeek-R1 and OpenAI o1-preview, struggle with tasks in LR$2$Bench.
  arXiv  Detail & Related papers  (2025-02-25T04:51:17Z)
• WILT: A Multi-Turn, Memorization-Robust Inductive Logic Benchmark for LLMs [0.8883751685905831]
  We introduce the Wason Inductive Logic Test (WILT), a simple yet challenging multi-turn reasoning benchmark designed to resist memorization. Our findings reveal that LLMs struggle with this task, exhibiting distinct strengths and weaknesses. Despite these variations, the best-performing model achieves only 28% accuracy, highlighting a significant gap in LLM performance on complex multi-turn reasoning tasks.
  arXiv  Detail & Related papers  (2024-10-14T18:29:13Z)
• TurtleBench: Evaluating Top Language Models via Real-World Yes/No Puzzles [2.8839090723566296]
  TurtleBench collects real user guesses from our online Turtle Soup Puzzle platform. TurtleBench includes 1,532 user guesses along with the correctness of guesses after annotation. We thoroughly evaluated nine of the most advanced Large Language Models available today.
  arXiv  Detail & Related papers  (2024-10-07T17:58:47Z)
• HELMET: How to Evaluate Long-Context Language Models Effectively and Thoroughly [34.205934899868346]
  We introduce HELMET, a comprehensive benchmark encompassing seven diverse, application-centric categories.<n>We find that synthetic tasks like NIAH do not reliably predict downstream performance.<n>While most LCLMs achieve perfect NIAH scores, open-source models significantly lag behind closed ones when tasks require full-context reasoning.
  arXiv  Detail & Related papers  (2024-10-03T17:20:11Z)
• Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines [74.42485647685272]
  We focus on Generative Masked Language Models (GMLMs) We train a model to fit conditional probabilities of the data distribution via masking, which are subsequently used as inputs to a Markov Chain to draw samples from the model. We adapt the T5 model for iteratively-refined parallel decoding, achieving 2-3x speedup in machine translation with minimal sacrifice in quality.
  arXiv  Detail & Related papers  (2024-07-22T18:00:00Z)
• LiveBench: A Challenging, Contamination-Limited LLM Benchmark [93.57775429120488]
  We release LiveBench, the first benchmark that contains frequently-updated questions from recent information sources.<n>We evaluate many prominent closed-source models, as well as dozens of open-source models ranging from 0.5B to 405B in size.<n>Questions are added and updated on a monthly basis, and we release new tasks and harder versions of tasks over time.
  arXiv  Detail & Related papers  (2024-06-27T16:47:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.