Newest papers

SPLASH: Contrastive Audit Pressure for Detecting and Localizing Security Hallucinations in Generated Code

Mirza Masfiqur Rahman, Ajwad Akil, Imtiaz Karim, Ashish Kundu, Ramana Rao Kompella, Elisa Bertino · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26001 · cited by 0 works (OpenAlex) · open-access PDF

Russian

Код, генерируемый большими языковыми моделями, часто проходит каждый функциональный тест, оставаясь неопределенным. Эти галлюцинации безопасности распространены, написаны с высокой степенью уверенности и повторяются по образцам, поэтому ни собственная неопределенность генератора, ни статические анализаторы надежно их не улавливают. Мы вводим SPLASH, который руководствуется новой мерой, называемой аудиторским давлением. Он контрастирует с вынужденными учителем лог-вероятностями для созданной программы при обзоре безопасности и обзоре стиля. Объединенные к заявлениям и программам, и основанные на вероятностной модели конкурирующих редактирований, аудиторское давление позволяет SPLASH выбирать безопасную программу среди кандидатурных программ модели и

Machine translation (Meta M2M100); the English original follows.

Abstract

Code generated by large language models often passes every functional test while remaining insecure. These security hallucinations are common, written with high confidence, and repeated across samples, so neither the generator's own uncertainty nor static analyzers reliably catch them. We introduce SPLASH, which is guided by a new measure called audit pressure. It contrasts teacher-forced log-probabilities for the generated program under a security review and a style review. Aggregated to statements and programs, and grounded in a probabilistic model of competing edits, audit pressure allows SPLASH to choose the secure program among a model's candidate programs and locate the vulnerable statement, with no model training, CWE-specific rules, or internal access to the generating model. On 63,684 programs generated by 12 models for 899 tasks, SPLASH picks secure code in 74% of cases, places the vulnerable statement first in 58% of insecure programs, and chooses better than a frontier LLM judge.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Code (set theory), Audit, Probabilistic logic, Computer security, Measure (data warehouse)

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26001, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange