paper-with-me

홈 › Papers

Fill-in-the-Blank: A Challenging Video Understanding Evaluation Framework

2021-11-16 · ACL ARR November 2021 11 · Anonymous

We propose fill-in-the-blanks as a video understanding evaluation framework. The task tests a model's understanding of a video by requiring the model to predict a masked noun phrase in the caption of the video, given the video and the surrounding text. To this end, we introduce a novel dataset consisting of 28,000 videos and fill-in-the-blank tests with multiple correct answers. The task and the dataset are challenging for the current state-of-the-art systems to solve. This task also does not share the weaknesses of the current state of the art language-informed video understanding tasks, namely: (1) video question answering using multiple-choice questions, where models perform relatively well because they exploit linguistic biases in the task formulation; and (2) video captioning, which relies on an open-ended evaluation framework that is often inaccurate because system answers may be perceived as incorrect if they differ in form from the ground truth.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Multiple-choiceQuestion AnsweringVideo CaptioningVideo Question AnsweringVideo Understanding

Similar Papers 제목 키워드 기반

FIBER: Fill-in-the-Blanks as a Challenging Video Understanding Evaluation Framework

2021-04-09 · ACL 2022 5 · Santiago Castro, Ruoyao Wang, Pingxuan Huang, Ian Stewart 외

We propose fill-in-the-blanks as a video understanding evaluation framework and introduce FIBER -- a novel dataset consisting of 28,000 videos and descriptions in support of this evaluation framework. The fill-in-the-bla…

Language ModellingMultiple-choiceQuestion AnsweringVideo Captioning+2

Video Fill in the Blank with Merging LSTMs

2016-10-13 · Amir Mazaheri, Dong Zhang, Mubarak Shah

Given a video and its incomplete textural description with missing words, the Video-Fill-in-the-Blank (ViFitB) task is to automatically find the missing word. The contextual information of the sentences are important to …

A dataset and exploration of models for understanding video data through fill-in-the-blank question-answering

2016-11-23 · CVPR 2017 7 · Tegan Maharaj, Nicolas Ballas, Anna Rohrbach, Aaron Courville 외

While deep convolutional neural networks frequently approach or exceed human-level performance at benchmark tasks involving static images, extending this success to moving images is not straightforward. Having models whi…

DescriptiveLanguage ModelingLanguage Modellingobject-detection+2

Video Fill In the Blank using LR/RL LSTMs with Spatial-Temporal Attentions

2017-04-15 · ICCV 2017 10 · Amir Mazaheri, Dong Zhang, Mubarak Shah

Given a video and a description sentence with one missing word (we call it the "source sentence"), Video-Fill-In-the-Blank (VFIB) problem is to find the missing word automatically. The contextual information of the sente…

Sentence

QUIET: A Multi-Blank Cascaded Story Cloze Benchmark for LLM Creative Generation Capability

2026-05-25 · Bo Zou, Chao Xu arxiv

Large language models (LLMs) face a dual challenge in creative capability evaluation: existing benchmarks (e.g., Story Cloze Test, HellaSwag) measure models' discriminative ability over narrative continuation using multi…

Logical ReasoningCloze Test