Skip to content
Open access

Critical Compilation Strictness for Certified Selective Hardening Against Compiler-Conditioned DNN Backdoors

Oct 2026 · Electronics · Vol 15, pp. 4804 · 0 citations

TL;DR

This work characterized this minimum by decision-violating transformed supports and instantiate an exact implicit hitting-set synthesis method with an executable witness oracle, which is the unique minimum decision-preserving policy over the 10,000 evaluated triggered inputs.

Abstract

Deep-learning compiler transformations can change security-sensitive inference decisions without changing trained model parameters. Critical compilation strictness measures the smallest set of enforceable source/reference-semantic sites whose protection excludes a prescribed decision violation for every legal joint compiler configuration on a declared input domain. We characterize this minimum by decision-violating transformed supports and instantiate an exact implicit hitting-set synthesis method with an executable witness oracle. On a five-site ConvNet evaluated using PyTorch 2.6.0+cpu and its Inductor backend, the triggered target-label proportion increases from 0.1078 under eager execution to 0.9996 under compilation; protecting only the feature stem is the unique minimum decision-preserving policy over the 10,000 evaluated triggered inputs. An executable oracle recovers this policy in seven protected-set queries (425.59 s total, including 320.21 s of first-use compilation/setup). A second architecture, VGG16-BN, also has a one-site minimum under Inductor. An independently trained ONNX Runtime 1.21.0 case yields the same unique minimum on a separate 5000-input policy-synthesis domain and records no eligible violations under that fixed policy across its 16 compatible configurations on the 10,000-input evaluation domain. In a controlled Inductor timing experiment, selective strictness reduces warmed batch latency by approximately 29% relative to all-strict execution. The exact guarantees apply to the specified finite input and compiler-configuration families; structural experiments separately exhibit higher-order interactions and differences between exact and heuristic selection.

Read PDF

Similar papers

Checking Leakage Witnesses versus Certifying Bounded Non-Leakage

When a language-model audit finds no leak, what is needed to certify non-leakage? We study guarantees over a declared prompt domain under an executable leakage criterion and decoding rule. For general bounded polynomial-time evaluators, a supplied leaking execution is polynomial-time checkable, while leak existence is...

Chao Feng, Burkhard Stiller · 0 citations
Review Oct 2026

CISB-Bench: An Auditable Source--IR Dataset of Compiler-Introduced Security Bugs

Compiler-introduced security bugs (CISBs) arise when an optimization, lowering, or instrumentation decision changes a security-relevant property of the generated program. They are difficult to study because their evidence is distributed across issue reports, reduced tests, historical configurations, and compiler artifa...

Saatvik Pradhan, Anuroop Saini, Pranav Attrey et al. · 0 citations
Book Open access Oct 2026

BanditFuzz: A Synergistic Differential Fuzzing Framework for Compilers

Compiler fuzzing is critical for ensuring the reliability of modern optimizing compilers. However, traditional generators and raw Large Language Model (LLM) outputs often yield syntactically invalid or low-value programs, incurring significant computational overhead. In this paper, we present BanditFuzz, an end-to-end...

Dong-Hui Li, Ying-Ying Li, Jin-Long Xu et al. · 0 citations
#artificial intelligence Book Open access Sep 2026

Semantic Prefix Oracles for LLM Decoding: Contracts and Differential Validation

Constrained decoding can enforce regular or context-free output formats, but many program-generation failures are semantic: scope, typing, and declaration effects depend on context. We present semantic grammar specifications, a declarative formalism that attaches such constraints to a context-free surface and executes...

Paul Kronlund-Drouault · 0 citations
#software testing Preprint Sep 2026

Verified Learning for Compiler Optimization: An LLM-Guided Architecture with Formal Control

This work fine-tune a code-centric LLM on transformations produced by Wyvern and embed Alive2 into a feedback loop that enforces semantic preservation for every generated rewrite, demonstrating that generative AI components can be safely integrated into compiler pipelines through deterministic validation and structured...

Devyani Singh, Rong Feng, Suman Saha · 0 citations
Sep 2026

Audit-First Recombination for Test-Time LLM Optimization Under Deterministic Verifiers.

Deterministic verifiers make test-time optimization of large language model (LLM) outputs practical, but modern search pipelines still waste budget on candidates that are structurally ineligible for semantic evaluation. Ineligibility includes parse failures, schema violations, and dependency breaks. These dead-on-arriv...

Rui Li, Shuang Cao · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.