REVIEW 5 cited by
Improving fine-grained understanding in image-text pre-training
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
Improving fine-grained understanding in image-text pre-training
read the original abstract
We introduce SPARse Fine-grained Contrastive Alignment (SPARC), a simple method for pretraining more fine-grained multimodal representations from image-text pairs. Given that multiple image patches often correspond to single words, we propose to learn a grouping of image patches for every token in the caption. To achieve this, we use a sparse similarity metric between image patches and language tokens and compute for each token a language-grouped vision embedding as the weighted average of patches. The token and language-grouped vision embeddings are then contrasted through a fine-grained sequence-wise loss that only depends on individual samples and does not require other batch samples as negatives. This enables more detailed information to be learned in a computationally inexpensive manner. SPARC combines this fine-grained loss with a contrastive loss between global image and text embeddings to learn representations that simultaneously encode global and local information. We thoroughly evaluate our proposed method and show improved performance over competing approaches both on image-level tasks relying on coarse-grained information, e.g. classification, as well as region-level tasks relying on fine-grained information, e.g. retrieval, object detection, and segmentation. Moreover, SPARC improves model faithfulness and captioning in foundational vision-language models.
Forward citations
Cited by 5 Pith papers
-
Improving CLIP Adaptation by Breaking Tail Alignment for Source-Free Cross-Domain Few-Shot Learning
Proposes Adaptive Tail-Head Alignment (ATHA) that breaks alignment for low-similarity 'tail tokens' in CLIP to boost source-free cross-domain few-shot learning.
-
Compared to What? Baselines and Metrics for Counterfactual Prompting
Counterfactual prompting effects on LLMs are often indistinguishable from those caused by meaning-preserving paraphrases, causing most previously reported demographic sensitivities to disappear under proper statistica...
-
All in One: A Unified Synthetic Data Pipeline for Multimodal Video Understanding
A unified synthetic data generation pipeline produces unlimited annotated multimodal video data across multiple tasks, enabling models trained mostly on synthetic data to generalize effectively to real-world video und...
-
Attention Grounded Enhancement for Visual Document Retrieval
AGREE boosts visual document retrieval by adding local relevance signals from MLLM attention maps to global document labels during retriever training.
-
Xray-Visual Models: Scaling Vision models on Industry Scale Data
A 2-billion-parameter vision encoder trained on 15B+ image-text and billions of video-hashtag pairs reports SOTA ImageNet linear-probe, Kinetics, and retrieval numbers, but relies on proprietary data and has several v...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.