用固定输入和轻量MLP提升视觉语言模型在产品评分预测中的效率与精度。
Bounded-Compute Multimodal Regression for Product-Rating Prediction
- 用轻量MLP替代语言模型头，结合固定图像和截断元数据实现确定性推理。
- 在1600万训练样本下验证相关性达0.39 PLCC，优于小规模训练集。
- 适合资源受限场景的高效多模态回归任务，可复现性强。
视觉语言模型（VLMs）在多模态质量评估中日益受到关注，但其默认依赖自回归文本生成和动态视觉处理，难以满足严格延迟预算下的标量回归需求。本文针对LoViF 2026高效VLM挑战，对SmolVLM2-256M-Video-Instruct进行受限计算适配。受近期多模态互动预测研究启发，采用基于特征的回归优于基于标记的评分生成，因此将语言建模头替换为由池化解码器状态驱动的轻量双层MLP，并通过固定384x384图像和截断元数据强制输入确定性。控制消融实验表明，静态全局图像处理略优于动态分块；训练样本从10万增至1600万时，验证相关性显著提升。在官方测试集上，228M参数模型取得0.39 PLCC和0.40 CES，为资源受限的多模态回归任务提供了强且可复现的基线。
原文摘要 · Abstract (English)
Vision-language models (VLMs) are increasingly attractive for multimodal quality assessment, but their default reliance on autoregressive text generation and dynamic visual processing is poorly matched to scalar regression under strict latency budgets. We present a bounded-compute adaptation of SmolVLM2-256M-Video-Instruct for product-rating prediction in the LoViF 2026 Efficient VLM challenge. Motivated by recent multimodal engagement-prediction results showing that feature-based regression can outperform token-based score generation, we replace the language-modeling head with a lightweight two-layer MLP fed by pooled decoder states, and we enforce deterministic inputs through fixed 384x384 images and truncated metadata. Across controlled ablations, static global image processing slightly outperforms dynamic tiling, and scaling from 100K to 16M training examples substantially improves validation correlation. Under the official held-out evaluation, our 228M-parameter model achieves 0.39 PLCC and 0.40 CES, providing a strong and reproducible baseline for resource-constrained multimodal regression.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。