paper-with-me

홈 › Papers

Tips and Tricks for Visual Question Answering: Learnings from the 2017 Challenge

2017-08-09 · CVPR 2018 6 · Damien Teney, Peter Anderson, Xiaodong He, Anton Van Den Hengel

This paper presents a state-of-the-art model for visual question answering (VQA), which won the first place in the 2017 VQA Challenge. VQA is a task of significant importance for research in artificial intelligence, given its multimodal nature, clear evaluation protocol, and potential real-world applications. The performance of deep neural networks for VQA is very dependent on choices of architectures and hyperparameters. To help further research in the area, we describe in detail our high-performing, though relatively simple model. Through a massive exploration of architectures and hyperparameters representing more than 3,000 GPU-hours, we identified tips and tricks that lead to its success, namely: sigmoid outputs, soft training targets, image features from bottom-up attention, gated tanh activations, output embeddings initialized using GloVe and Google Images, large mini-batches, and smart shuffling of training data. We provide a detailed analysis of their impact on performance to assist others in making an appropriate selection.

📄 PDF Abstract BibTeX arXiv:1708.02711

Code (10)

SinghJasdeep/Attention-on-Attention-for-VQA pytorch
VincentYing/Attention-on-Attention-for-VQA pytorch
feifengwhu/question_attention pytorch
hengyuan-hu/bottom-up-attention-vqa pytorch
peteanderson80/bottom-up-attention
shailzajolly/Understanding-yesno-and-nonyesno-samples-VQA pytorch
shailzajolly/icdar_vqa pytorch
snagiri/ECE285_Jarvis_ProjectA pytorch
thilinicooray/Bottom-up-vqa pytorch
yangdsh/VQA-BUTD-demo

Tasks

GPUVisual Question AnsweringVisual Question Answering (VQA)

Methods 이 논문이 사용한 방법론

GloVe GloVe Embeddings are a type of word embedding that encode the co-occurrence probability ratio between two words as vector differences. GloVe uses a weighted least squares…

Similar Papers 제목 키워드 기반

A guide to the dataset explosion in QA, NLI, and commonsense reasoning

2020-12-01 · COLING 2020 8 · Anna Rogers, Anna Rumshisky

Question answering, natural language inference and commonsense reasoning are increasingly popular as general NLP system benchmarks, driving both modeling and dataset work. Only for question answering we already have over…

Natural Language InferenceQuestion Answering

TIPS Over Tricks: Simple Prompts for Effective Zero-shot Anomaly Detection

2026-02-03 · Alireza Salehi, Ehsan Karami, Sepehr Noey, Sahand Noey 외 arxiv

Anomaly detection identifies departures from expected behavior in safety-critical settings. When target-domain normal data are unavailable, zero-shot anomaly detection (ZSAD) leverages vision-language models (VLMs). Howe…

Anomaly Detection

Trajectory-Informed Memory Generation for Self-Improving Agent Systems

2026-03-11 · Gaodan Fang, Vatche Isahagian, K. R. Jayaram, Ritesh Kumar 외 arxiv

LLM-powered agents face a persistent challenge: learning from their execution experiences to improve future performance. While agents can successfully complete many tasks, they often repeat inefficient patterns, fail to …

The Answer is at your Fingertips: Improving Passage Retrieval for Web Question Answering with Search Behavior Data

2013-10-01 · EMNLP 2013 10 · Mikhail Ageev, Dmitry Lagun, Eugene Agichtein
Passage RetrievalQuestion AnsweringRetrieval

TIPS: Turn-Level Information-Potential Reward Shaping for Search-Augmented LLMs

2026-03-11 · Yutao Xie, Nathaniel Thomas, Nicklas Hansen, Yang Fu 외 arxiv

Search-augmented large language models (LLMs) trained with reinforcement learning (RL) have achieved strong results on open-domain question answering (QA), but training still remains a significant challenge. The optimiza…

Open-Domain Question AnsweringReinforcement Learning