LLaVAl vs. Fine-Tuned SLMs: Robustness Gaps on Adversarial Code Benchmarks

Abstract

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: What is the robustness gap between LLaVul and fine-tuned SLMs on adversarially perturbed code samples from the Devign benchmark, measured in terms of accuracy drop under input obfuscation. Detecting toxic content using language models is crucial yet challenging. While substantial progress has been made in English, toxicity detection in French remains underdeveloped, primarily due to the lack of culturally relevant, human-annotated, large-scale datasets. 13 claims were extracted from source literature; 1 was independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 4.2/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research Question

What is the robustness gap between LLaVul and fine-tuned SLMs on adversarially perturbed code samples from the Devign benchmark, measured in terms of accuracy drop under input obfuscation?

Verification Level

Descriptive public verification status only; aggregate claim counts are public, but individual claim records are not exposed here.

Quality Tier

Descriptive public triage only; this tier does not alter current publication or DOI behavior.

Quality Dimensions

Automated triage signals derived from public fields; not human peer review or independent validation.

Correction Record

Public corrections are additive records. Current status does not claim the synthesis is error-free.

Provenance