Don't Judge a Book by its Cover: Testing LLMs' Robustness Under Logical Obfuscation

• URL: http://arxiv.org/abs/2602.01132v1
• Date: Sun, 01 Feb 2026 10:04:22 GMT
• Title: Don't Judge a Book by its Cover: Testing LLMs' Robustness Under Logical Obfuscation
• Abstract summary: We present LogiQAte, a first-of-its-kind diagnostic benchmark with 1,108 questions across four reasoning tasks.<n>We find that obfuscation severely degrades zero-shot performance, with performance dropping on average by 47% for GPT-4o, 27% for GPT-5, and 22% for reasoning model, o4-mini.
• Score: 3.9997332571742934
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Tasks such as solving arithmetic equations, evaluating truth tables, and completing syllogisms are handled well by large language models (LLMs) in their standard form, but they often fail when the same problems are posed in logically equivalent yet obfuscated formats. To study this vulnerability, we introduce Logifus, a structure-preserving logical obfuscation framework, and, utilizing this, we present LogiQAte, a first-of-its-kind diagnostic benchmark with 1,108 questions across four reasoning tasks: (i) Obfus FOL (first-order logic entailment under equivalence-preserving rewrites), (ii) Obfus Blood Relation (family-graph entailment under indirect relational chains), (iii) Obfus Number Series (pattern induction under symbolic substitutions), and (iv) Obfus Direction Sense (navigation reasoning under altered directions and reference frames). Across all the tasks, evaluating six state-of-the-art models, we find that obfuscation severely degrades zero-shot performance, with performance dropping on average by 47% for GPT-4o, 27% for GPT-5, and 22% for reasoning model, o4-mini. Our findings reveal that current LLMs parse questions without deep understanding, highlighting the urgency of building models that genuinely comprehend and preserve meaning beyond surface form.

Related papers

• HOLMES: Evaluating Higher-Order Logical Reasoning in LLMs [37.82259837085897]
  We introduce HOLMES (Higher-Order Logic Meets real-world Explainable reasoning), the first real-world benchmark for higher-order symbolic reasoning in LLMs.<n>Built on higher-order logic, HOLMES pairs natural-language problems with HOL formalizations, ground-truth answers, verifiable reasoning traces, and fine-grained controllable reasoning factors across law and finance.<n>Experiments show that current LLMs still struggle on HOLMES, with an average accuracy of only 50.64% and the best model reaching 59.54%.
  arXiv  Detail & Related papers  (2026-06-22T12:23:54Z)
• DeonticBench: A Benchmark for Reasoning over Rules [52.69517904415795]
  DEONTICBENCH is a benchmark of 6,232 tasks across U.S. federal taxes, airline baggage policies, U.S. immigration administration, and U.S. state housing law.<n>It is a benchmark for studying context-grounded rule reasoning in real-world domains under both symbolic and non-symbolic settings.
  arXiv  Detail & Related papers  (2026-04-06T05:41:02Z)
• From Early Encoding to Late Suppression: Interpreting LLMs on Character Counting Tasks [49.57538588967748]
  Large language models (LLMs) exhibit failures on elementary symbolic tasks such as character counting in a word, despite excelling on complex benchmarks.<n>We uncover a consistent phenomenon across modern architectures, including LLaMA, Qwen, and Gemma.<n>We show that symbolic reasoning failures in LLMs are not due to missing representations or insufficient scale, but arise from structured interference within the model's computation graph.
  arXiv  Detail & Related papers  (2026-04-01T11:40:12Z)
• Task Complexity Matters: An Empirical Study of Reasoning in LLMs for Sentiment Analysis [2.1036545320600095]
  Large language models (LLMs) with reasoning capabilities have fueled a compelling narrative that reasoning universally improves performance across language tasks.<n>We test this claim through a comprehensive evaluation of 504 configurations across seven model families.<n>Our findings reveal that reasoning effectiveness is strongly task-dependent, challenging prevailing assumptions.
  arXiv  Detail & Related papers  (2026-02-27T14:49:05Z)
• From Abstract to Contextual: What LLMs Still Cannot Do in Mathematics [79.81905350372067]
  We study gap through contextual mathematical reasoning.<n>We introduce ContextMATH, a benchmark that repurposes AIME and MATH-500 problems into two contextual settings.<n>Open-source models decline by 13 and 34 points on SG and CS, while proprietary models drop by 13 and 20.
  arXiv  Detail & Related papers  (2026-01-30T14:56:04Z)
• The Semantic Trap: Do Fine-tuned LLMs Learn Vulnerability Root Cause or Just Functional Pattern? [14.472036099680961]
  We propose TrapEval, a comprehensive evaluation framework designed to disentangle vulnerability root cause from functional pattern.<n>We fine-tune five representative state-of-the-art LLMs across three model families and evaluate them under cross-dataset testing, semantic-preservings, and varying degrees of semantic gap measured by CodeBLEU.<n>Our findings serve as a wake-up call: high benchmark scores on traditional datasets may be illusory, masking the model's inability to understand the true causal logic of vulnerabilities.
  arXiv  Detail & Related papers  (2026-01-30T07:19:17Z)
• From Harm to Help: Turning Reasoning In-Context Demos into Assets for Reasoning LMs [58.02809208460186]
  We revisit this paradox using high-quality traces from DeepSeek-R1 as demonstrations.<n>We find that adding more exemplars consistently degrades accuracy, even when demonstrations are optimal.<n>We introduce Insight-to-solve (I2S), a sequential test-time procedure that turns demonstrations into explicit, reusable insights.
  arXiv  Detail & Related papers  (2025-09-27T08:59:31Z)
• Chain-of-Code Collapse: Reasoning Failures in LLMs via Adversarial Prompting in Code Generation [0.3495246564946556]
  Large Language Models (LLMs) have achieved remarkable success in tasks requiring complex reasoning.<n>Do these models truly reason, or do they merely exploit shallow statistical patterns?<n>We introduce Chain-of-Code Collapse, where we investigate the robustness of reasoning LLMs by introducing a suite of semantically faithful yet adversarially structured prompt perturbations.
  arXiv  Detail & Related papers  (2025-06-08T02:43:46Z)
• Benchmarking Abstract and Reasoning Abilities Through A Theoretical Perspective [59.7140089198992]
  We develop a mathematic framework that defines abstract reasoning as the ability to extract essential patterns.<n>We introduce two novel complementary metrics: (scoreGamma) measures basic reasoning accuracy, while (scoreDelta) quantifies a model's reliance on specific symbols.
  arXiv  Detail & Related papers  (2025-05-28T09:02:45Z)
• Beyond Chains of Thought: Benchmarking Latent-Space Reasoning Abilities in Large Language Models [0.0]
  Large language models (LLMs) can perform reasoning computations both internally within their latent space and externally.<n>This study introduces a benchmark designed to quantify model-internal reasoning in different domains.
  arXiv  Detail & Related papers  (2025-04-14T18:15:27Z)
• Large Language Models Are Better Logical Fallacy Reasoners with Counterargument, Explanation, and Goal-Aware Prompt Formulation [2.4073494101588273]
  This study presents a novel and effective prompt formulation approach for logical fallacy detection.<n>Our method enriches input text incorporating implicit contextual information, which we query for validity within the context of the argument.<n>We evaluate our approach across multiple datasets from 5 domains, covering 29 distinct fallacy types.
  arXiv  Detail & Related papers  (2025-03-30T08:41:09Z)
• Self-Discover: Large Language Models Self-Compose Reasoning Structures [136.48389510481758]
  We introduce SELF-DISCOVER, a framework for self-discovering task-intrinsic reasoning structures. SELF-DISCOVER substantially improves GPT-4 and PaLM 2's performance on challenging reasoning benchmarks. We show that the self-discovered reasoning structures are universally applicable across model families.
  arXiv  Detail & Related papers  (2024-02-06T01:13:53Z)
• LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models [63.14196038655506]
  We introduce LogicAsker, a novel approach for evaluating and enhancing the logical reasoning capabilities of large language models (LLMs) Our methodology reveals significant gaps in LLMs' learning of logical rules, with identified reasoning failures ranging from 29% to 90% across different models. We leverage these findings to construct targeted demonstration examples and fine-tune data, notably enhancing logical reasoning in models like GPT-4o by up to 5%.
  arXiv  Detail & Related papers  (2024-01-01T13:53:53Z)
• Language Models can be Logical Solvers [99.40649402395725]
  We introduce LoGiPT, a novel language model that directly emulates the reasoning processes of logical solvers. LoGiPT is fine-tuned on a newly constructed instruction-tuning dataset derived from revealing and refining the invisible reasoning process of deductive solvers.
  arXiv  Detail & Related papers  (2023-11-10T16:23:50Z)
• Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning [101.26814728062065]
  Large Language Models (LLMs) have shown human-like reasoning abilities but still struggle with complex logical problems. This paper introduces a novel framework, Logic-LM, which integrates LLMs with symbolic solvers to improve logical problem-solving.
  arXiv  Detail & Related papers  (2023-05-20T22:25:38Z)

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.