English

Silkie: Preference Distillation for Large Visual Language Models

Computer Vision and Pattern Recognition 2023-12-19 v1 Computation and Language

Abstract

This paper explores preference distillation for large vision language models (LVLMs), improving their ability to generate helpful and faithful responses anchoring the visual context. We first build a vision-language feedback (VLFeedback) dataset utilizing AI annotation. Specifically, responses are generated by models sampled from 12 LVLMs, conditioned on multi-modal instructions sourced from various datasets. We adopt GPT-4V to assess the generated outputs regarding helpfulness, visual faithfulness, and ethical considerations. Furthermore, the preference supervision is distilled into Qwen-VL-Chat through the direct preference optimization (DPO) method. The resulting model Silkie, achieves 6.9% and 9.5% relative improvement on the MME benchmark regarding the perception and cognition capabilities, respectively. Silkie also demonstrates reduced hallucination by setting a new state-of-the-art score of 3.02 on the MMHal-Bench benchmark. Further analysis shows that DPO with our VLFeedback dataset mainly boosts the fine-grained perception and complex cognition abilities of LVLMs, leading to more comprehensive improvements compared to human-annotated preference datasets.

Keywords

Cite

@article{arxiv.2312.10665,
  title  = {Silkie: Preference Distillation for Large Visual Language Models},
  author = {Lei Li and Zhihui Xie and Mukai Li and Shunian Chen and Peiyi Wang and Liang Chen and Yazheng Yang and Benyou Wang and Lingpeng Kong},
  journal= {arXiv preprint arXiv:2312.10665},
  year   = {2023}
}

Comments

Project page: https://vlf-silkie.github.io

R2 v1 2026-06-28T13:53:50.930Z