DPO Training with Human-Ranked Paraphrases Enhances Code Generation Robustness

This report synthesises findings from 6 peer-reviewed papers addressing the following research question: How does DPO training with human-ranked paraphrase data affect the robustness of code generation models against syntax-preserving adversarial perturbations on HumanEval Pro compared to standard SFT. In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and. 7 claims were extracted from source literature; 7 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 8.5/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: How does DPO training with human-ranked paraphrase data affect the robustness of code generation models against syntax-preserving adversarial perturbations on HumanEval Pro compared to standard SFT baselines?

Autonomous literature synthesis. Automated review score: 8.5/10. Full text and citation available at Assignee Research.