REVIEW 4 cited by
Pythia v0.1: the Winning Entry to the VQA Challenge 2018
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
Pythia v0.1: the Winning Entry to the VQA Challenge 2018
read the original abstract
This document describes Pythia v0.1, the winning entry from Facebook AI Research (FAIR)'s A-STAR team to the VQA Challenge 2018. Our starting point is a modular re-implementation of the bottom-up top-down (up-down) model. We demonstrate that by making subtle but important changes to the model architecture and the learning rate schedule, fine-tuning image features, and adding data augmentation, we can significantly improve the performance of the up-down model on VQA v2.0 dataset -- from 65.67% to 70.22%. Furthermore, by using a diverse ensemble of models trained with different features and on different datasets, we are able to significantly improve over the 'standard' way of ensembling (i.e. same model with different random seeds) by 1.31%. Overall, we achieve 72.27% on the test-std split of the VQA v2.0 dataset. Our code in its entirety (training, evaluation, data-augmentation, ensembling) and pre-trained models are publicly available at: https://github.com/facebookresearch/pythia
Forward citations
Cited by 4 Pith papers
-
VisualBERT: A Simple and Performant Baseline for Vision and Language
VisualBERT is a Transformer model that implicitly aligns text and image regions through self-attention and achieves competitive or superior results on VQA, VCR, NLVR2, and Flickr30K after pre-training on captions.
-
MIND: Multi-rationale INtegrated Discriminative Reasoning Framework for Multi-modal Large Models
MIND improves multimodal reasoning by training on diverse correct and deliberately wrong rationales with two-stage correction and contrastive alignment, reporting SOTA on ScienceQA, A-OKVQA, and M3CoT.
-
Adversarial Regularization for Visual Question Answering: Strengths, Shortcomings, and Side Effects
Adversarial regularization improves VQA performance on out-of-domain bias tests but introduces unstable gradients, reduced in-domain accuracy, and over-reliance on visual cues at the expense of linguistic information.
-
DDVT: Dynamic Dual-level Vision Transformer Fusion Network for Answer Grounding in Visual Question Answering
DDVT, a Swin+BERT answer-grounding network with question-guided dynamic mask-resolution selection and cross-modal multi-scale aggregation, reports 65.3 IoU on VizWizGround and 43.47 on VQS, roughly 12 and 5.5 points a...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.