Right Answer, Wrong Method: Shortcut Hacking Misleads the Evaluation of LLM Reasoning on Frontier Science Benchmarks

• URL: http://arxiv.org/abs/2608.02442v1
• Date: Mon, 03 Aug 2026 16:21:56 GMT
• Title: Right Answer, Wrong Method: Shortcut Hacking Misleads the Evaluation of LLM Reasoning on Frontier Science Benchmarks
• Abstract summary: We identify Solution Hacking, a failure mode in which an LLM reaches the correct answer through invalid shortcuts.<n>We develop expert-inspired anti-hacking strategies, including an automatic judge and a test-time instruction.
• Score: 3.3243946853975346
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Scientific reasoning benchmarks typically evaluate large language models (LLMs) using final-answer accuracy. However, a correct answer does not necessarily demonstrate the reasoning capability targeted by the problem. We identify Solution Hacking, a failure mode in which an LLM reaches the correct answer through invalid shortcuts, such as numerical search, enumeration, guessing, or answer-first verification, without providing a valid task-targeted derivation. We systematically analyze this phenomenon across difficulty levels, scientific domains, and frontier models. Solution hacking increases sharply with benchmark difficulty, from 2.2\% on common problems to 28.3\% on Olympiad-level problems and 37.4\% on HLE. Moreover, 8.2\%-44.1\% of answers credited as correct across frontier models are identified as hacked solutions. We further develop expert-inspired anti-hacking strategies, including an automatic judge and a test-time instruction. The results show that suppressing shortcut behavior substantially reduces reported accuracy while having a smaller effect on correct and non-hacked accuracy. These findings reveal that answer-only evaluation can overestimate the scientific reasoning capabilities of frontier LLMs.

Related papers

• Deep Interaction: An Efficient Human-AI Interaction Method for Large Reasoning Models [16.90017128904664]
  We propose an efficient human intervention mechanism for precisely correcting reasoning errors in large language models (LLMs)<n>Our approach enables direct editing of the original response, allowing erroneous parts to be corrected while preserving accurate reasoning steps.<n>Our method achieves over a 25% improvement in correction success rate and reduces token usage by approximately 40% on STEM tasks reasoning.
  arXiv  Detail & Related papers  (2026-07-15T17:16:43Z)
• Verifying Large Language Models' Reasoning Paths via Correlation Matrix Rank [71.09032766271493]
  Large language models (LLMs) are prone to errors and hallucinations.<n>How to check their outputs effectively and efficiently has become a critical problem in their applications.
  arXiv  Detail & Related papers  (2025-10-28T11:01:10Z)
• Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs [57.10533368622962]
  Self-correction of large language models (LLMs) emerges as a critical component for enhancing their reasoning performance.<n>This study introduces CorrectBench, a benchmark developed to evaluate the effectiveness of self-correction strategies.<n>Our findings reveal that: 1) Self-correction methods can improve accuracy, especially for complex reasoning tasks; 2) Mixing different self-correction strategies yields further improvements, though it reduces efficiency; and 3) Reasoning LLMs (e.g., DeepSeek-R1) have limited optimization under additional self-correction methods and have high time costs.
  arXiv  Detail & Related papers  (2025-10-17T02:40:19Z)
• Hard2Verify: A Step-Level Verification Benchmark for Open-Ended Frontier Math [80.46254366870447]
  We introduce Hard2Verify, a step-level verification benchmark produced with over 500 hours of human labor.<n>We evaluate 29 generative critics and process reward models, demonstrating that, beyond a few standouts, open-source verifiers lag closed source models.
  arXiv  Detail & Related papers  (2025-10-15T16:50:54Z)
• Right Is Not Enough: The Pitfalls of Outcome Supervision in Training LLMs for Math Reasoning [35.142294794883455]
  We introduce MathOlympiadEval, a new dataset with fine-grained annotations, which reveals a significant gap between LLMs' answer correctness and their low process correctness.<n>Existing automated methods like LLM-as-a-judge struggle to reliably detect these reasoning flaws.<n>We propose ParaStepVerifier, a novel methodology for meticulous, step-by-step verification of mathematical solutions.
  arXiv  Detail & Related papers  (2025-06-07T17:54:56Z)
• 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)
• Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing [59.405145971637204]
  We propose a novel preference learning framework called eRror-Injected Self-Editing (RISE)<n>RISE injects predefined subtle errors into pivotal tokens in reasoning or steps to construct hard pairs for error mitigation.<n>Experiments validate the effectiveness of RISE, with preference learning on Qwen2-7B-Instruct yielding notable improvements of 3.0% on GSM8K and 7.9% on MATH with only 4.5K training samples.
  arXiv  Detail & Related papers  (2024-10-09T07:43:38Z)
• Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verification [52.095460362197336]
  Large language models (LLMs) struggle with consistent and accurate reasoning. LLMs are trained primarily on correct solutions, reducing their ability to detect and learn from errors. We propose a novel collaborative method integrating Chain-of-Thought (CoT) and Program-of-Thought (PoT) solutions for verification.
  arXiv  Detail & Related papers  (2024-10-05T05:21:48Z)
• SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning [55.76083560152823]
  SelfCheck is a general-purpose zero-shot verification schema for recognizing errors in step-by-step reasoning. We test SelfCheck on three datasets (GSM8K, MathQA, and MATH) and find that it successfully recognizes errors and, in turn, increases final answer accuracies.
  arXiv  Detail & Related papers  (2023-08-01T10:31:36Z)

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.