PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models
This work addresses a key bottleneck in VLP models for vision-language tasks, offering a solution that enhances performance on position-sensitive applications while maintaining efficiency, though it is incremental in building on existing VLP frameworks.
The paper tackles the challenge of explicit object modeling in detector-free vision-language pre-training (VLP) models, which are efficient but lack position sensitivity for tasks like referring expression comprehension; it introduces PEVL, a method that integrates object positions into a unified language modeling framework, achieving state-of-the-art performance on position-sensitive tasks and improving results on position-insensitive ones with grounded inputs.
Vision-language pre-training (VLP) has shown impressive performance on a wide range of cross-modal tasks, where VLP models without reliance on object detectors are becoming the mainstream due to their superior computation efficiency and competitive performance. However, the removal of object detectors also deprives the capability of VLP models in explicit object modeling, which is essential to various position-sensitive vision-language (VL) tasks, such as referring expression comprehension and visual commonsense reasoning. To address the challenge, we introduce PEVL that enhances the pre-training and prompt tuning of VLP models with explicit object position modeling. Specifically, PEVL reformulates discretized object positions and language in a unified language modeling framework, which facilitates explicit VL alignment during pre-training, and also enables flexible prompt tuning for various downstream tasks. We show that PEVL enables state-of-the-art performance of detector-free VLP models on position-sensitive tasks such as referring expression comprehension and phrase grounding, and also improves the performance on position-insensitive tasks with grounded inputs. We make the data and code for this paper publicly available at https://github.com/thunlp/PEVL.