Generating Complex Code Analyzers from Natural Language Questions

• URL: http://arxiv.org/abs/2605.09304v1
• Date: Sun, 10 May 2026 04:02:10 GMT
• Title: Generating Complex Code Analyzers from Natural Language Questions
• Abstract summary: We present Merlin, a new system for answering free-form questions that require analytical reasoning about code.<n>We develop a RAG-based iterative query-generation approach and a novel self-test technique.
• Score: 23.315299132854438
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many software development tasks, such as implementing features and fixing bugs, begin with developers posing questions about a codebase. However, answering questions about codebases that span millions of lines of code across thousands of files is non-trivial. Standard tools like grep cannot answer questions requiring semantic or inter-procedural reasoning, and large language models (LLMs) struggle with large codebases due to resource and context constraints. In this paper, we present Merlin, a new system for answering free-form questions that require analytical reasoning about code. Merlin integrates an LLM with CodeQL, a program analysis framework that supports expressive queries over large codebases. We face two principal challenges in the design of such systems: First, program analysis queries are diverse and semantically complex; as a result, even syntactically well-formed queries frequently produce degenerate/empty results. Furthermore, relatively few CodeQL queries are available online, limiting the out-of-the-box effectiveness of LLMs as CodeQL query generators. We address these challenges by developing a RAG-based iterative query-generation approach and a novel self-test technique. Our query debugging technique builds on the idea of assistive queries, which generate concrete witnesses that expose and explain semantic flaws in candidate queries. We evaluate Merlin through both experimental and user studies. Over a set of natural language questions derived from common bug-finding tasks, Merlin discovered not only the majority of software issues reported by other approaches, but also issues that would have otherwise remained undetected. Through a within-subject user study, we found that access to Merlin increased task accuracy by an average of 3.8* and simultaneously reduced the time for programmers to complete all tasks by 31%.

Related papers

• Beyond Caption-Based Queries for Video Moment Retrieval [60.31221310786333]
  We investigate degradation of VMR methods when trained on caption-based queries but evaluated on search queries.<n>We introduce three benchmarks by modifying the textual queries in three public VMR datasets.<n>Our approach improves performance on search queries by up to 14.82% mAP_m, and up to 21.83% mAP_m on multi-moment search queries.
  arXiv  Detail & Related papers  (2026-03-02T20:06:41Z)
• AlignCoder: Aligning Retrieval with Target Intent for Repository-Level Code Completion [55.21541958868449]
  We propose AlignCoder, a repository-level code completion framework.<n>Our framework generates an enhanced query that bridges the semantic gap between the initial query and the target code.<n>We employ reinforcement learning to train an AlignRetriever that learns to leverage inference information in the enhanced query for more accurate retrieval.
  arXiv  Detail & Related papers  (2026-01-27T15:23:14Z)
• Text-to-SPARQL Goes Beyond English: Multilingual Question Answering Over Knowledge Graphs through Human-Inspired Reasoning [51.203811759364925]
  mKGQAgent breaks down the task of converting natural language questions into SPARQL queries into modular, interpretable subtasks.<n> Evaluated on the DBpedia- and Corporate-based KGQA benchmarks within the Text2SPARQL challenge 2025, our approach took first place among the other participants.
  arXiv  Detail & Related papers  (2025-07-22T19:23:03Z)
• The benefits of query-based KGQA systems for complex and temporal questions in LLM era [55.20230501807337]
  Large language models excel in question-answering (QA) yet still struggle with multi-hop reasoning and temporal questions.<n> Query-based knowledge graph QA (KGQA) offers a modular alternative by generating executable queries instead of direct answers.<n>We explore multi-stage query-based framework for WikiData QA, proposing multi-stage approach that enhances performance on challenging multi-hop and temporal benchmarks.
  arXiv  Detail & Related papers  (2025-07-16T06:41:03Z)
• CodeAssistBench (CAB): Dataset & Benchmarking for Multi-turn Chat-Based Code Assistance [18.886738819470086]
  We introduce CodeAssistBench (CAB), the first benchmark framework for evaluating multi-turn programming assistance.<n>Unlike existing programming Q&A benchmarks, CAB automatically generates scalable datasets from question-related GitHub issues.<n>Using this framework, we constructed a test set of 3,286 real-world programming questions across 231 repositories.
  arXiv  Detail & Related papers  (2025-07-14T17:19:00Z)
• Q${}^2$Forge: Minting Competency Questions and SPARQL Queries for Question-Answering Over Knowledge Graphs [6.6757601046766135]
  The SPARQL query language is the standard method to access knowledge graphs (KGs)<n>Best practices recommend to document KGs with competency questions and example queries.<n>Q$2$Forge addresses the challenge of generating new competency questions for a KG and corresponding SPARQL queries.
  arXiv  Detail & Related papers  (2025-05-19T13:26:51Z)
• Steering Large Language Models between Code Execution and Textual Reasoning [22.279107036500083]
  Textual reasoning has inherent limitations in solving tasks with challenges in math, logics, optimization, and searching.<n>OpenAI GPT Code Interpreter and multi-agent frameworks such as AutoGen have demonstrated remarkable proficiency of integrating code generation and execution.<n>We propose three methods to better steer LLM code/text generation and achieve a notable improvement.
  arXiv  Detail & Related papers  (2024-10-04T15:44:47Z)
• What's Wrong with Your Code Generated by Large Language Models? An Extensive Study [92.62952504133926]
  This study evaluated the performance of three leading closed-source LLMs and six popular open-source LLMs on three commonly used benchmarks.<n>We developed a taxonomy of bugs for incorrect codes and analyzed the root cause for common bug types.<n>We propose a novel training-free iterative method that introduces self-critique, enabling LLMs to critique and correct their generated code.
  arXiv  Detail & Related papers  (2024-07-08T17:27:17Z)
• InfiBench: Evaluating the Question-Answering Capabilities of Code Large Language Models [56.723509505549536]
  InfiBench is the first large-scale freeform question-answering (QA) benchmark for code to our knowledge. It comprises 234 carefully selected high-quality Stack Overflow questions that span across 15 programming languages. We conduct a systematic evaluation for over 100 latest code LLMs on InfiBench, leading to a series of novel and insightful findings.
  arXiv  Detail & Related papers  (2024-03-11T02:06:30Z)
• Allies: Prompting Large Language Model with Beam Search [107.38790111856761]
  In this work, we propose a novel method called ALLIES. Given an input query, ALLIES leverages LLMs to iteratively generate new queries related to the original query. By iteratively refining and expanding the scope of the original query, ALLIES captures and utilizes hidden knowledge that may not be directly through retrieval.
  arXiv  Detail & Related papers  (2023-05-24T06:16:44Z)

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.