Order Matters! An Empirical Study on Large Language Models' Input Order Bias in Software Fault Localization

• URL: http://arxiv.org/abs/2412.18750v4
• Date: Fri, 26 Sep 2025 19:33:14 GMT
• Title: Order Matters! An Empirical Study on Large Language Models' Input Order Bias in Software Fault Localization
• Abstract summary: This study investigates the impact of input order and context size on Large Language Models (LLMs) performance.<n>Our results indicate a significant bias in order; Top-1 FL accuracy in Java projects drops from 57% to 20%, while in Python projects, it decreases from 38% to approximately 3% when we reverse the code order.<n>Breaking down inputs into smaller contexts helps reduce this bias, narrowing the performance gap in FL from 22% to 6% and then to just 1% on both benchmarks.
• Score: 17.03610523319531
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) show great promise in software engineering tasks like Fault Localization (FL) and Automatic Program Repair (APR). This study investigates the impact of input order and context size on LLM performance in FL, a crucial step for many downstream software engineering tasks. We test different orders for methods using Kendall Tau distances, including "perfect" (where ground truths come first) and "worst" (where ground truths come last), using two benchmarks that consist of both Java and Python projects. Our results indicate a significant bias in order; Top-1 FL accuracy in Java projects drops from 57% to 20%, while in Python projects, it decreases from 38% to approximately 3% when we reverse the code order. Breaking down inputs into smaller contexts helps reduce this bias, narrowing the performance gap in FL from 22% to 6% and then to just 1% on both benchmarks. We then investigated whether the bias in order was caused by data leakage by renaming the method names with more meaningful alternatives. Our findings indicated that the trend remained consistent, suggesting that the bias was not due to data leakage. We also look at ordering methods based on traditional FL techniques and metrics. Ordering using DepGraph's ranking achieves 48% Top-1 accuracy, which is better than more straightforward ordering approaches like CallGraphDFS. These findings underscore the importance of how we structure inputs, manage contexts, and choose ordering methods to improve LLM performance in FL and other software engineering tasks.

Related papers

• SieveFL: Hierarchical Runtime-Aware Pruning for Scalable LLM-Based Fault Localization [0.0]
  Automated fault localization requires connecting an observed test failure to the responsible method across thousands of candidates.<n>We present SieveFL, a five-stage hierarchical framework that resolves this tension through aggressive pre-LLM filtering.
  arXiv  Detail & Related papers  (2026-05-13T13:16:41Z)
• How Low Can You Go? The Data-Light SE Challenge [4.282746516699565]
  Much of software engineering (SE) research assumes that progress depends on massive datasets and CPU-intensives.<n>Counter-evidence presented in this paper suggests otherwise, including software configuration and performance tuning, cloud and systems optimization, project and process-level decision modeling, behavioral analytics, financial risk modeling, project health prediction, reinforcement learning tasks, sales forecasting, and software testing.<n>Our results highlight that some SE tasks may be better served by lightweight approaches that demand fewer labels and far less computation.
  arXiv  Detail & Related papers  (2025-12-15T16:49:50Z)
• 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)
• Teaching LLM to Reason: Reinforcement Learning from Algorithmic Problems without Code [76.80306464249217]
  We propose TeaR, which aims at teaching LLMs to reason better.<n>TeaR leverages careful data curation and reinforcement learning to guide models in discovering optimal reasoning paths through code-related tasks.<n>We conduct extensive experiments using two base models and three long-CoT distillation models, with model sizes ranging from 1.5 billion to 32 billion parameters, and across 17 benchmarks spanning Math, Knowledge, Code, and Logical Reasoning.
  arXiv  Detail & Related papers  (2025-07-10T07:34:05Z)
• Efficiency-Effectiveness Reranking FLOPs for LLM-based Rerankers [17.658542084440082]
  Large Language Models (LLMs) have been applied to reranking tasks in information retrieval, achieving strong performance.<n>Existing studies evaluate the efficiency of LLM-based rerankers using proxy metrics such as latency, the number of forward passes, input tokens, and output tokens.<n>We propose RPP (ranking metrics per PetaFLOP), measuring how much ranking quality (e.g., NDCG or MRR) a method achieves per PetaFLOP, and QPP (queries per PetaFLOP), measuring how many queries can be processed per PetaFLOP.
  arXiv  Detail & Related papers  (2025-07-08T17:56:28Z)
• MIB: A Mechanistic Interpretability Benchmark [77.35046700898326]
  We propose MIB, a Mechanistic Interpretability Benchmark, with two tracks spanning four tasks and five models.<n>Using MIB, we find that attribution and mask optimization methods perform best on circuit localization.<n>For causal variable localization, we find that the supervised DAS method performs best, while SAE features are not better than neurons.
  arXiv  Detail & Related papers  (2025-04-17T17:55:45Z)
• SmartFL: Semantics Based Probabilistic Fault Localization [15.481820762877897]
  Testing-based fault localization has been a research focus in software engineering in the past decades. It is crucial to model program semantics in fault localization approaches. Our key idea is: by modeling only the correctness of program values but not their full semantics, a balance could be reached between effectiveness and scalability.
  arXiv  Detail & Related papers  (2025-03-29T21:00:51Z)
• PromptOptMe: Error-Aware Prompt Compression for LLM-based MT Evaluation Metrics [21.23509339665165]
  We propose a prompt optimization approach that uses a smaller, fine-tuned language model to compress input data for evaluation prompt.<n>Our results show a $2.37times$ reduction in token usage without any loss in evaluation quality.
  arXiv  Detail & Related papers  (2024-12-20T18:08:02Z)
• RAZOR: Sharpening Knowledge by Cutting Bias with Unsupervised Text Rewriting [16.633948320306832]
  biases prevalent in manually constructed datasets can introduce spurious correlations between tokens and labels. Existing debiasing methods often rely on prior knowledge of specific dataset biases. We propose RAZOR, a novel, unsupervised, and data-focused debiasing approach based on text rewriting for shortcut mitigation.
  arXiv  Detail & Related papers  (2024-12-10T17:02:58Z)
• Enhancing Fault Localization Through Ordered Code Analysis with LLM Agents and Self-Reflection [8.22737389683156]
  Large Language Models (LLMs) offer promising improvements in fault localization by enhancing code comprehension and reasoning. We introduce LLM4FL, a novel LLM-agent-based fault localization approach that integrates SBFL rankings with a divide-and-conquer strategy. Our results demonstrate that LLM4FL outperforms AutoFL by 19.27% in Top-1 accuracy and surpasses state-of-the-art supervised techniques such as DeepFL and Grace.
  arXiv  Detail & Related papers  (2024-09-20T16:47:34Z)
• Impact of Large Language Models of Code on Fault Localization [2.936007114555107]
  We propose a simple but effective sequence generation approach for fine-tuning large language models of code for FL tasks. Specifically, we fine-tune representative encoder, encoder-decoder, and decoder-based 13 LLMCs for FL tasks. Experimental results show that LLMCs fine-tuned with our approach successfully pinpoint error positions in 50.6%, 64.2%, and 72.3% of 1,291 methods in Defects4J for Top-2/3/5 prediction.
  arXiv  Detail & Related papers  (2024-08-19T02:36:07Z)
• Eliminating Position Bias of Language Models: A Mechanistic Approach [119.34143323054143]
  Position bias has proven to be a prevalent issue of modern language models (LMs) Our mechanistic analysis attributes the position bias to two components employed in nearly all state-of-the-art LMs: causal attention and relative positional encodings. By eliminating position bias, models achieve better performance and reliability in downstream tasks, including LM-as-a-judge, retrieval-augmented QA, molecule generation, and math reasoning.
  arXiv  Detail & Related papers  (2024-07-01T09:06:57Z)
• Not Everything is All You Need: Toward Low-Redundant Optimization for Large Language Model Alignment [126.34547428473968]
  Large language models (LLMs) are still struggling in aligning with human preference in complex tasks and scenarios. We propose a low-redundant alignment method named textbfALLO, focusing on optimizing the most related neurons with the most useful supervised signals. Experimental results on 10 datasets have shown the effectiveness of ALLO.
  arXiv  Detail & Related papers  (2024-06-18T13:34:40Z)
• Revisiting Unnaturalness for Automated Program Repair in the Era of Large Language Models [9.454475517867817]
  We propose a patch-naturalness measurement, entropy-delta, to improve the efficiency of template-based repair techniques. Our proposed method can rank correct patches more effectively than state-of-the-art machine learning tools.
  arXiv  Detail & Related papers  (2024-04-23T17:12:45Z)
• MinPrompt: Graph-based Minimal Prompt Data Augmentation for Few-shot Question Answering [64.6741991162092]
  We present MinPrompt, a minimal data augmentation framework for open-domain question answering. We transform the raw text into a graph structure to build connections between different factual sentences. We then apply graph algorithms to identify the minimal set of sentences needed to cover the most information in the raw text. We generate QA pairs based on the identified sentence subset and train the model on the selected sentences to obtain the final model.
  arXiv  Detail & Related papers  (2023-10-08T04:44:36Z)
• Large Language Models for Test-Free Fault Localization [11.080712737595174]
  We propose a language model based fault localization approach that locates buggy lines of code without any test coverage information. We fine-tune language models with 350 million, 6 billion, and 16 billion parameters on small, manually curated corpora of buggy programs. Our empirical evaluation shows that LLMAO improves the Top-1 results over the state-of-the-art machine learning fault localization (MLFL) baselines by 2.3%-54.4%, and Top-5 results by 14.4%-35.6%.
  arXiv  Detail & Related papers  (2023-10-03T01:26:39Z)
• Benchmarking Causal Study to Interpret Large Language Models for Source Code [6.301373791541809]
  This paper introduces a benchmarking strategy named Galeras comprised of curated testbeds for three SE tasks. We illustrate the insights of our benchmarking strategy by conducting a case study on the performance of ChatGPT under distinct prompt engineering methods.
  arXiv  Detail & Related papers  (2023-08-23T20:32:12Z)
• GRACE: Discriminator-Guided Chain-of-Thought Reasoning [75.35436025709049]
  We propose Guiding chain-of-thought ReAsoning with a CorrectnEss Discriminator (GRACE) to steer the decoding process towards producing correct reasoning steps. GRACE employs a discriminator trained with a contrastive loss over correct and incorrect steps, which is used during decoding to score next-step candidates.
  arXiv  Detail & Related papers  (2023-05-24T09:16:51Z)
• SatLM: Satisfiability-Aided Language Models Using Declarative Prompting [68.40726892904286]
  We propose a new satisfiability-aided language modeling (SatLM) approach for improving the reasoning capabilities of large language models (LLMs) We use an LLM to generate a declarative task specification rather than an imperative program and leverage an off-the-shelf automated theorem prover to derive the final answer. We evaluate SATLM on 8 different datasets and show that it consistently outperforms program-aided LMs in the imperative paradigm.
  arXiv  Detail & Related papers  (2023-05-16T17:55:51Z)
• Re-Evaluating LiDAR Scene Flow for Autonomous Driving [80.37947791534985]
  Popular benchmarks for self-supervised LiDAR scene flow have unrealistic rates of dynamic motion, unrealistic correspondences, and unrealistic sampling patterns. We evaluate a suite of top methods on a suite of real-world datasets. We show that despite the emphasis placed on learning, most performance gains are caused by pre- and post-processing steps.
  arXiv  Detail & Related papers  (2023-04-04T22:45:50Z)
• Communication-Efficient Robust Federated Learning with Noisy Labels [144.31995882209932]
  Federated learning (FL) is a promising privacy-preserving machine learning paradigm over distributed located data. We propose a learning-based reweighting approach to mitigate the effect of noisy labels in FL. Our approach has shown superior performance on several real-world datasets compared to various baselines.
  arXiv  Detail & Related papers  (2022-06-11T16:21:17Z)
• Bi-level Alignment for Cross-Domain Crowd Counting [113.78303285148041]
  Current methods rely on external data for training an auxiliary task or apply an expensive coarse-to-fine estimation. We develop a new adversarial learning based method, which is simple and efficient to apply. We evaluate our approach on five real-world crowd counting benchmarks, where we outperform existing approaches by a large margin.
  arXiv  Detail & Related papers  (2022-05-12T02:23:25Z)
• Local Learning Matters: Rethinking Data Heterogeneity in Federated Learning [61.488646649045215]
  Federated learning (FL) is a promising strategy for performing privacy-preserving, distributed learning with a network of clients (i.e., edge devices)
  arXiv  Detail & Related papers  (2021-11-28T19:03:39Z)
• Fast Uncertainty Quantification for Deep Object Pose Estimation [91.09217713805337]
  Deep learning-based object pose estimators are often unreliable and overconfident. In this work, we propose a simple, efficient, and plug-and-play UQ method for 6-DoF object pose estimation.
  arXiv  Detail & Related papers  (2020-11-16T06:51:55Z)

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.