REVIEW 2 cited by
Distilled Dual-Encoder Model for Vision-Language Understanding
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We propose a cross-modal attention distillation framework to train a dual-encoder model for vision-language understanding tasks, such as visual reasoning and visual question answering. Dual-encoder models have a faster inference speed than fusion-encoder models and enable the pre-computation of images and text during inference. However, the shallow interaction module used in dual-encoder models is insufficient to handle complex vision-language understanding tasks. In order to learn deep interactions of images and text, we introduce cross-modal attention distillation, which uses the image-to-text and text-to-image attention distributions of a fusion-encoder model to guide the training of our dual-encoder model. In addition, we show that applying the cross-modal attention distillation for both pre-training and fine-tuning stages achieves further improvements. Experimental results demonstrate that the distilled dual-encoder model achieves competitive performance for visual reasoning, visual entailment and visual question answering tasks while enjoying a much faster inference speed than fusion-encoder models. Our code and models will be publicly available at https://github.com/kugwzk/Distilled-DualEncoder.
Forward citations
Cited by 2 Pith papers
-
EffiVLM-BENCH: A Comprehensive Benchmark for Evaluating Training-Free Acceleration in Large Vision-Language Models
EffiVLM-Bench is a benchmark study showing token compression is task- and model-dependent, KV cache methods are more loyal, and parameter compression preserves accuracy better at typical ratios.
-
HIT Model: A Hierarchical Interaction-Enhanced Two-Tower Model for Pre-Ranking Systems
HIT augments the two-tower pre-ranking architecture with class-conditional generators and multi-head MaxSim matching, reporting consistent AUC gains on three public datasets and +1.66% GMV in a Tencent online A/B test.
Discussion (0). Sign in to comment.