# VLRM: Vision-Language Models act as Reward Models for Image Captioning

Source: [VLRM: Vision-Language Models act as Reward Models for Image Captioning](<https://www.sota2.com/research/paper/68fa72b91e209131cea58279>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:32:50.348Z

Authors: Maksim Dzabraev, Alexander Kunitsyn, Andrei Ivaniuta

Published: 2024-04-02T12:57:22.000Z

## Abstract

In this work, we present an unsupervised method for enhancing an image captioning model (in our case, BLIP2) using reinforcement learning and vision-language models like CLIP and BLIP2-ITM as reward models. The RL-tuned model is able to generate longer and more comprehensive descriptions. Our model reaches impressive 0.90 R@1 CLIP Recall score on MS-COCO Carpathy Test Split.   Weights are available at https://huggingface.co/sashakunitsyn/vlrm-blip2-opt-2.7b.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2404.01911>)

[PDF](<https://arxiv.org/pdf/2404.01911.pdf>)