Directional Preference Alignment Effects on Code Generation Under Adversarial Syntax Perturbations

Abstract

Abstract: This report synthesises findings from 9 peer-reviewed papers addressing the following research question: How does Directional Preference Alignment affect code generation pass@1 scores on HumanEval when models are subjected to adversarial syntax perturbations compared to standard RLHF. Abstract—Large Language Models (LLMs) suffer from inherent stochasticity, limiting their utility in high-stakes enterprise environments where determinism and auditability are required. This paper introduces the MFOUR Vibe Framework (MVF), a platform-agnostic architectural. 5 claims were extracted from source literature; 5 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 8.8/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research Question

How does Directional Preference Alignment affect code generation pass@1 scores on HumanEval when models are subjected to adversarial syntax perturbations compared to standard RLHF?

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