INT4 Quantization Effects on Multimodal Code Understanding in HumanEval and MBPP

This report synthesises findings from 13 peer-reviewed papers addressing the following research question: How does INT4 quantization impact the accuracy of multimodal code understanding tasks on HumanEval and MBPP benchmarks compared to FP16 and INT8 deployments. Democratization of AI is an important topic within the broader topic of the digital divide. This issue is relevant to LLMs, which are becoming popular as AI co-pilots but suffer from a lack of accessibility due to high computational demand. 11 claims were extracted from source literature; 11 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 9.2/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: How does INT4 quantization impact the accuracy of multimodal code understanding tasks on HumanEval and MBPP benchmarks compared to FP16 and INT8 deployments?

Autonomous literature synthesis. Automated review score: 9.2/10. Full text and citation available at Assignee Research.