Lemma Discovery in Agentic Program Verification

• URL: http://arxiv.org/abs/2603.22114v1
• Date: Mon, 23 Mar 2026 15:42:07 GMT
• Title: Lemma Discovery in Agentic Program Verification
• Abstract summary: Deductive verification provides strong correctness guarantees for code by extracting verification conditions (VCs) and writing formal proofs for them.<n>The expertise-intensive task of VC proving is the main bottleneck in this process, and has been partly automated owing to recent advances in Large Language Model (LLM) agents.<n>We argue that VC proving for program verification is more than a purely mathematical task, and benefits considerably from program comprehension.
• Score: 12.551445518827768
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deductive verification provides strong correctness guarantees for code by extracting verification conditions (VCs) and writing formal proofs for them. The expertise-intensive task of VC proving is the main bottleneck in this process, and has been partly automated owing to recent advances in Large Language Model (LLM) agents. However, existing proof agents are not able to discover helper lemmas - auxiliary lemmas that aid in proving - and thus fall short as programs grow in size and complexity. In this paper, we argue that VC proving for program verification is more than a purely mathematical task, and benefits considerably from program comprehension. Our key insight is that human-proof engineers often discover and apply helper lemmas based on their understanding of the program semantics, which are not directly reflected in the VCs produced by VC generators. Inspired by this insight, we propose an LLM agent, LemmaNet, that discovers helper lemmas in two ways. Specifically, the agent first synthesizes lemmas offline by directly analyzing the source code and specifications, and then relating this semantic understanding to the mechanical, verbose encoding produced by VC generators. As the proof unfolds, LemmaNet then adapts existing helper lemmas online to accommodate evolving proof states, enabling the agent to effectively discharge complex VCs on-the-fly. We evaluate LemmaNet on SV-COMP and established real-world subjects, including modules of the Linux kernel, Contiki OS, standard C++ library, and X.509 parser. Our experimental results demonstrate that LemmaNet significantly outperforms state-of-the-art approaches, highlighting the importance of program comprehension-aided lemma discovery in agentic program verification.

Related papers

• The Imitation Game: When LLMs Learn to Reason Like Programs via Code-Centric Reasoning Data Synthesis [50.473217152006875]
  MIMIC is a framework that leverages executable code as a rigorous medium for reasoning data synthesis.<n>MIMIC transforms algorithms into verifiable reasoning trajectories through narrative fusion, code-guided test synthesis, and dynamic code instrumentation.<n>Our method significantly elevates accuracy across general reasoning, complex mathematical benchmarks, and fine-grained deterministic tasks.
  arXiv  Detail & Related papers  (2026-09-13T17:04:16Z)
• On Automated and Explainable Provenance of AI-Generated Code [56.818202904805446]
  We argue that the next generation of CodeGenAI tools must be built on a foundation of explainable provenance.<n>We describe a research program combining large-scale empirical studies with post-hoc causal and interpretability techniques.
  arXiv  Detail & Related papers  (2026-08-03T14:50:31Z)
• ClarifyCodeBench: Evaluating LLMs on Clarifying Ambiguous Requirements for Code Generation [54.788849448970154]
  We introduce ClarifyCodeBench, a novel interactive benchmark for evaluating Large Language Models' capability in resolving requirement ambiguity.<n>We conduct a systematic evaluation of six state-of-the-art LLMs using ClarifyCodeBench.<n>Our empirical results yield three critical insights: 1) Capability Decoupling: Strong code generation performance does not inherently translate to effective requirement clarification; 2) The Reasoning Paradox: While increased computational thinking enhances code correctness, it yields marginal gains in identifying ambiguities.
  arXiv  Detail & Related papers  (2026-07-01T09:58:59Z)
• AgentV-RL: Scaling Reward Modeling with Agentic Verifier [63.55502685076245]
  Verifiers have been demonstrated to enhance LLM reasoning via test-time scaling (TTS)<n>We propose Agentic Verifier, a framework that transforms reward modeling into a multi-turn, tool-augmented deliberative process.<n>We show that Agentic Verifier yields consistent performance gains under both parallel and sequential TTS.
  arXiv  Detail & Related papers  (2026-04-17T12:27:36Z)
• Veri-Sure: A Contract-Aware Multi-Agent Framework with Temporal Tracing and Formal Verification for Correct RTL Code Generation [4.723302382132762]
  silicon-grade correctness remains bottlenecked by: (i) limited test coverage and reliability of simulation-centric evaluation, (ii) regressions and repair hallucinations, and (iii) semantic drift as intent is reinterpreted across agent handoffs.<n>We propose Veri-Sure, a multi-agent framework that establishes a design contract to align agents' intent and uses a patching mechanism guided by static dependency slicing to perform precise, localized repairs.
  arXiv  Detail & Related papers  (2026-01-27T16:10:23Z)
• Neural Theorem Proving for Verification Conditions: A Real-World Benchmark [9.350519191460018]
  This work introduces Neural Theorem Proving for Verification Conditions (NTP4VC), presenting the first real-world multi-language benchmark for this task.<n>We evaluate large language models (LLMs), both general-purpose and those fine-tuned for theorem proving, on NTP4VC.
  arXiv  Detail & Related papers  (2026-01-26T20:37:11Z)
• BRIDGE: Building Representations In Domain Guided Program Verification [67.36686119518441]
  BRIDGE decomposes verification into three interconnected domains: Code, Specifications, and Proofs.<n>We show that this approach substantially improves both accuracy and efficiency beyond standard error feedback methods.
  arXiv  Detail & Related papers  (2025-11-26T06:39:19Z)
• Agentic Program Verification [14.684859166069012]
  We present a first Large Language Models agent, AutoRocq, for conducting program verification.<n>Unlike past works, which rely on extensive training of LLMs on proof examples, our agent learns on-the-fly and improves the proof via an iterative refinement loop.<n>In this way, our proof construction involves autonomous collaboration between the proof agent and the theorem prover.
  arXiv  Detail & Related papers  (2025-11-21T15:51:48Z)
• Dissect-and-Restore: AI-based Code Verification with Transient Refactoring [1.2883590530210827]
  We present Prometheus, a novel AI-assisted system that facilitates automated code verification with current AI capabilities.<n>Prometheus guides the proof search through structured decomposition of complex lemmas into smaller, verifiable sub-lemmas.<n>This approach successfully verifies 86% of tasks in our curated dataset, compared to 68% for the baseline.
  arXiv  Detail & Related papers  (2025-10-29T11:23:50Z)
• What's in a Proof? Analyzing Expert Proof-Writing Processes in F* and Verus [2.8003002159083237]
  We conduct a user study involving the collection and analysis of fine-grained source code telemetry from eight experts working with two languages.<n>Results reveal interesting trends and patterns about how experts reason about proofs and key challenges encountered during the proof development process.<n>We translate these findings into concrete design guidance for AI proof assistants.
  arXiv  Detail & Related papers  (2025-08-01T22:16:30Z)
• Re:Form -- Reducing Human Priors in Scalable Formal Software Verification with RL in LLMs: A Preliminary Study on Dafny [78.1575956773948]
  Large Language Models (LLMs) trained with Reinforcement Learning (RL) face a significant challenge: their verification processes are neither reliable nor scalable.<n>A promising yet largely uncharted alternative is formal language-based reasoning.<n>Grounding LLMs in rigorous formal systems where generative models operate in formal language spaces (e.g., Dafny) enables the automatic and mathematically provable verification of their reasoning processes and outcomes.
  arXiv  Detail & Related papers  (2025-07-22T08:13:01Z)
• Do AI models help produce verified bug fixes? [62.985237003585674]
  Large Language Models are used to produce corrections to software bugs.<n>This paper investigates how programmers use Large Language Models to complement their own skills.<n>The results are a first step towards a proper role for AI and LLMs in providing guaranteed-correct fixes to program bugs.
  arXiv  Detail & Related papers  (2025-07-21T17:30:16Z)
• Training Language Models to Generate Quality Code with Program Analysis Feedback [66.0854002147103]
  Code generation with large language models (LLMs) is increasingly adopted in production but fails to ensure code quality.<n>We propose REAL, a reinforcement learning framework that incentivizes LLMs to generate production-quality code.
  arXiv  Detail & Related papers  (2025-05-28T17:57:47Z)

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.