arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.00151cs.AIcs.CLcs.LG
收藏

DVD:一个视频接地对话多步推理诊断数据集

DVD: A Diagnostic Dataset for Multi-step Reasoning in Video Grounded Dialogue

Singapore Management University(新加坡管理大学);Institute for Infocomm Research, A*STAR(新加坡科技研究局资讯通信研究院);Facebook(脸书)

Hung Le, Chinnadhurai Sankar, Seungwhan Moon, Ahmad Beirami, Alborz Geramifard, Satwik Kottur

摘要:

基于视频的对话系统需要理解对话和视频,对话包含从一个回合到另一个回合的语义依赖关系,视频包含空间和时间场景变化的视觉线索。构建这样的对话系统是一个具有挑战性的问题,涉及到视觉和语言输入的各种推理类型。现有的基准没有足够的注释来彻底分析对话系统并孤立地理解它们的能力和局限性。这些基准也没有明确地设计成最小化模型在没有实际推理的情况下可以利用的偏差。为了解决这些局限性,在本文中,我们提出了DVD,一个诊断数据集的视频接地对话。该数据集被设计为包含最小的偏差,并对视频时空空间上的不同类型的推理进行了详细的注释。对话是在多个问题轮上合成的,每个问题轮都注入了一组跨轮语义关系。我们使用DVD来分析现有的方法,提供有趣的见解,他们的能力和局限性。总的来说,DVD是从$11k$ CATER合成视频构建的,并且包含每个视频的$10$回合对话的$10$实例,从而产生多个$100k$对话和$1M$问答对。我们的代码和数据集可在https://github.com/facebookresearch/DVDialogues上公开获取。

英文摘要:

A video-grounded dialogue system is required to understand both dialogue, which contains semantic dependencies from turn to turn, and video, which contains visual cues of spatial and temporal scene variations. Building such dialogue systems is a challenging problem, involving various reasoning types on both visual and language inputs. Existing benchmarks do not have enough annotations to thoroughly analyze dialogue systems and understand their capabilities and limitations in isolation. These benchmarks are also not explicitly designed to minimise biases that models can exploit without actual reasoning. To address these limitations, in this paper, we present DVD, a Diagnostic Dataset for Video-grounded Dialogues. The dataset is designed to contain minimal biases and has detailed annotations for the different types of reasoning over the spatio-temporal space of video. Dialogues are synthesized over multiple question turns, each of which is injected with a set of cross-turn semantic relationships. We use DVD to analyze existing approaches, providing interesting insights into their abilities and limitations. In total, DVD is built from $11k$ CATER synthetic videos and contains $10$ instances of $10$-round dialogues for each video, resulting in more than $100k$ dialogues and $1M$ question-answer pairs. Our code and dataset are publicly available at https://github.com/facebookresearch/DVDialogues.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑