KL-Constraint Hyperparameter Effects on RLHF and DPO Alignment in Corrupted Multimodal Benchmarks

Abstract

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: What is the impact of varying KL-constraint hyperparameters on the alignment performance of RLHF versus DPO when evaluated on corrupted image-text pairs from multimodal benchmarks like LLaVA-Bench. Aligning language models with human preferences through reinforcement learning from human feedback is crucial for their safe and effective deployment. The human preference is typically represented through comparison where one response is chosen over another for a given prompt. 11 claims were extracted from source literature; 0 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 2.8/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research Question

What is the impact of varying KL-constraint hyperparameters on the alignment performance of RLHF versus DPO when evaluated on corrupted image-text pairs from multimodal benchmarks like LLaVA-Bench?

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