Hierarchical Evaluation of Software Design Capabilities of Large Language Models of Code

• URL: http://arxiv.org/abs/2511.20933v1
• Date: Tue, 25 Nov 2025 23:50:00 GMT
• Title: Hierarchical Evaluation of Software Design Capabilities of Large Language Models of Code
• Authors: Mootez Saad, Boqi Chen, José Antonio Hernández López, Dániel Varró, Tushar Sharma,
• Abstract summary: Large language models (LLMs) are increasingly adopted in software engineering domain, yet robustness of their grasp on core design concepts remains unclear.<n>We generate poorly designed software fragments under various levels of guidance.<n> Reasoning about coupling proves brittle; performance collapses in noisy, open-ended scenarios.<n> Reasoning-trace analysis confirms these failure modes, revealing textitcognitive shortcutting for coupling versus a more exhaustive (yet still failing) analysis for cohesion.
• Score: 7.897548449569687
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) are being increasingly adopted in the software engineering domain, yet the robustness of their grasp on core software design concepts remains unclear. We conduct an empirical study to systematically evaluate their understanding of cohesion (intra-module) and coupling (inter-module). We programmatically generate poorly designed code fragments and test the DeepSeek-R1 model family ($14$B, $32$B, $70$B) under varying levels of guidance, from simple \textit{Verification} to \textit{Guided} and \textit{Open-ended Generation}, while varying contextual noise by injecting distractor elements. While models exhibit a solid baseline understanding of both concepts in ideal conditions, their practical knowledge is fragile and highly asymmetrical. Reasoning about coupling proves brittle; performance collapses in noisy, open-ended scenarios, with F1 scores dropping by over $50\%$. In contrast, the models' analysis of cohesion is remarkably robust to internal noise in guided tasks, showing little performance degradation. However, this resilience also fails when all guidance is removed. Reasoning-trace analysis confirms these failure modes, revealing \textit{cognitive shortcutting} for coupling versus a more exhaustive (yet still failing) analysis for cohesion. To summarize, while LLMs can provide reliable assistance for recognizing design flaws, their ability to reason autonomously in noisy, realistic contexts is limited, highlighting the critical need for more scalable and robust program understanding capabilities.

Related papers

• OMG-Agent: Toward Robust Missing Modality Generation with Decoupled Coarse-to-Fine Agentic Workflows [9.617220633655716]
  We present textbfunderlineOmni-textbfunderlineModality textbfunderlineGeneration Agent (textbfOMG-Agent)
  arXiv  Detail & Related papers  (2026-02-04T02:25:40Z)
• CoT-Seg: Rethinking Segmentation with Chain-of-Thought Reasoning and Self-Correction [50.67483317563736]
  This paper aims to explore a system that can think step-by-step, look up information if needed, generate results, self-evaluate its own results, and refine the results.<n>We introduce CoT-Seg, a training-free framework that rethinks reasoning segmentation by combining chain-of-thought reasoning with self-correction.
  arXiv  Detail & Related papers  (2026-01-24T11:41:54Z)
• CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs [53.199517625701475]
  CoG is a training-free framework inspired by Dual-Process Theory that mimics the interplay between intuition and deliberation.<n>CoG significantly outperforms state-of-the-art approaches in both accuracy and efficiency.
  arXiv  Detail & Related papers  (2026-01-16T07:27:40Z)
• Robust-R1: Degradation-Aware Reasoning for Robust Visual Understanding [54.05243949024302]
  Existing robust MLLMs rely on implicit training/adaptation that focuses solely on visual encoder generalization.<n>We propose Robust-R1, a novel framework that explicitly models visual degradations through structured reasoning chains.<n>Our approach integrates: (i) supervised fine-tuning for degradation-aware reasoning foundations, (ii) reward-driven alignment for accurately perceiving degradation parameters, and (iii) dynamic reasoning depth scaling adapted to degradation intensity.
  arXiv  Detail & Related papers  (2025-12-19T12:56:17Z)
• SEAL: Self-Evolving Agentic Learning for Conversational Question Answering over Knowledge Graphs [28.59157823781425]
  SEAL is a novel two-stage semantic parsing framework grounded in self-evolving agentic learning.<n> SEAL achieves state-of-the-art performance, especially in multi-hop reasoning, comparison, and aggregation tasks.<n>The results validate notable gains in both structural accuracy and computational efficiency.
  arXiv  Detail & Related papers  (2025-12-04T14:52:30Z)
• RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Benchmark [71.3555284685426]
  We introduce RealUnify, a benchmark designed to evaluate bidirectional capability synergy.<n>RealUnify comprises 1,000 meticulously human-annotated instances spanning 10 categories and 32 subtasks.<n>We find that current unified models still struggle to achieve effective synergy, indicating that architectural unification alone is insufficient.
  arXiv  Detail & Related papers  (2025-09-29T15:07:28Z)
• Sycophancy Mitigation Through Reinforcement Learning with Uncertainty-Aware Adaptive Reasoning Trajectories [58.988535279557546]
  We introduce textbf sycophancy Mitigation through Adaptive Reasoning Trajectories.<n>We show that SMART significantly reduces sycophantic behavior while preserving strong performance on out-of-distribution inputs.
  arXiv  Detail & Related papers  (2025-09-20T17:09:14Z)
• RationAnomaly: Log Anomaly Detection with Rationality via Chain-of-Thought and Reinforcement Learning [27.235259453535537]
  RationAnomaly is a novel framework that enhances log anomaly detection by synergizing Chain-of-Thought fine-tuning with reinforcement learning.<n>We have released the corresponding resources, including code and datasets.
  arXiv  Detail & Related papers  (2025-09-18T07:35:58Z)
• Understanding Software Engineering Agents: A Study of Thought-Action-Result Trajectories [17.975121612118752]
  Large Language Model (LLM)-based agents are increasingly employed to automate complex software engineering tasks.<n>We present a large-scale empirical study of the thought-action-result trajectories of three state-of-the-art LLM-based agents.<n>We identify key trajectory characteristics, such as counts and token consumption, recurring action sequences, and the semantic coherence of thoughts, actions, and their results.
  arXiv  Detail & Related papers  (2025-06-23T16:34:52Z)
• PixelThink: Towards Efficient Chain-of-Pixel Reasoning [70.32510083790069]
  PixelThink is a simple yet effective scheme that integrates externally estimated task difficulty and internally measured model uncertainty.<n>It learns to compress reasoning length in accordance with scene complexity and predictive confidence.<n> Experimental results demonstrate that the proposed approach improves both reasoning efficiency and overall segmentation performance.
  arXiv  Detail & Related papers  (2025-05-29T17:55:49Z)
• Towards a Science of Causal Interpretability in Deep Learning for Software Engineering [0.32634122554914]
  dissertation addresses achieving causal interpretability in Deep Learning for Software Engineering (DL4SE)<n> dissertation introduces DoCode, a novel post hoc interpretability method for Neural Code Models (NCMs)<n>DoCode uses causal inference to provide programming language-oriented explanations of model predictions.
  arXiv  Detail & Related papers  (2025-05-21T02:13:11Z)
• Thinking Longer, Not Larger: Enhancing Software Engineering Agents via Scaling Test-Time Compute [61.00662702026523]
  We propose a unified Test-Time Compute scaling framework that leverages increased inference-time instead of larger models.<n>Our framework incorporates two complementary strategies: internal TTC and external TTC.<n>We demonstrate our textbf32B model achieves a 46% issue resolution rate, surpassing significantly larger models such as DeepSeek R1 671B and OpenAI o1.
  arXiv  Detail & Related papers  (2025-03-31T07:31:32Z)
• Reasoning with LLMs for Zero-Shot Vulnerability Detection [0.9208007322096533]
  We present textbfVulnSage, a comprehensive evaluation framework and a curated dataset from diverse, large-scale open-source system software projects.<n>The framework supports multi-granular analysis across function, file, and inter-function levels.<n>It employs four diverse zero-shot prompt strategies: Baseline, Chain-of-context, Think, and Think & verify.
  arXiv  Detail & Related papers  (2025-03-22T23:59:17Z)
• Causality can systematically address the monsters under the bench(marks) [64.36592889550431]
  Benchmarks are plagued by various biases, artifacts, or leakage.<n>Models may behave unreliably due to poorly explored failure modes.<n> causality offers an ideal framework to systematically address these challenges.
  arXiv  Detail & Related papers  (2025-02-07T17:01:37Z)

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.