Related papers: YTCommentVerse: A Multi-Category Multi-Lingual You…
Responding with multi-modal content has been recognized as an essential capability for an intelligent conversational agent. In this paper, we introduce the MMDialog dataset to better facilitate multi-modal conversation. MMDialog is composed…
This paper contributes to the ongoing discussions on the scholarly access to social media data, discussing a case where this access is barred despite its value for understanding and countering online disinformation and despite the absence…
Today, comprehensive evaluation of large-scale machine learning models is possible thanks to the open datasets produced using crowdsourcing, such as SQuAD, MS COCO, ImageNet, SuperGLUE, etc. These datasets capture objective responses,…
While human annotations play a crucial role in language technologies, annotator subjectivity has long been overlooked in data collection. Recent studies that have critically examined this issue are often situated in the Western context, and…
This paper is focused on the computational analysis of collective discourse, a collective behavior seen in non-expert content contributions in online social media. We collect and analyze a wide range of real-world collective discourse…
Vlogs provide a rich public source of data in a novel setting. This paper examined the continuous sentiment styles employed in 27,333 vlogs using a dynamic intra-textual approach to sentiment analysis. Using unsupervised clustering, we…
Recently, the Metaverse is becoming increasingly attractive, with millions of users accessing the many available virtual worlds. However, how do users find the one Metaverse which best fits their current interests? So far, the search…
Sentiment analysis is one of the most fundamental tasks in Natural Language Processing. Popular languages like English, Arabic, Russian, Mandarin, and also Indian languages such as Hindi, Bengali, Tamil have seen a significant amount of…
YouTube is central to contemporary mass media. However, the official YouTube API does not provide access to the full set of creators or creator metadata on the platform. This lack of basic visibility into the YouTube ecosystem hinders…
To address the bottleneck of accurate user intent interpretation within the current video generation community, we present Any2Caption, a novel framework for controllable video generation under any condition. The key idea is to decouple…
Podcasts provide highly diverse content to a massive listener base through a unique on-demand modality. However, limited data has prevented large-scale computational analysis of the podcast ecosystem. To fill this gap, we introduce a…
Live video commenting is popular on video media platforms, as it can create a chatting atmosphere and provide supplementary information for users while watching videos. Automatically generating live video comments can improve user…
Sentiment Analysis is currently a vital area of research. With the advancement in the use of the internet, the creation of social media, websites, blogs, opinions, ratings, etc. has increased rapidly. People express their feedback and…
Sentiment analysis using big data from YouTube videos metadata can be conducted to analyze public opinions on various political figures who represent political parties. This is possible because YouTube has become one of the platforms for…
Toxic language remains an ongoing challenge on social media platforms, presenting significant issues for users and communities. This paper provides a cross-topic and cross-lingual analysis of toxicity in Reddit conversations. We collect 1.5…
Sarcasm detection is a significant challenge in sentiment analysis, particularly due to its nature of conveying opinions where the intended meaning deviates from the literal expression. This challenge is heightened in social media contexts…
We present 3MASSIV, a multilingual, multimodal and multi-aspect, expertly-annotated dataset of diverse short videos extracted from short-video social media platform - Moj. 3MASSIV comprises of 50k short videos (20 seconds average duration)…
We introduce TalkVerse, a large-scale, open corpus for single-person, audio-driven talking video generation designed to enable fair, reproducible comparison across methods. While current state-of-the-art systems rely on closed data or…
Over one in five adults in the US lives with a mental illness. In the face of a shortage of mental health professionals and offline resources, online short-form video content has grown to serve as a crucial conduit for disseminating mental…
Sequence-to-sequence models have recently gained the state of the art performance in summarization. However, not too many large-scale high-quality datasets are available and almost all the available ones are mainly news articles with…