Topics of Interest

Authors are invited to submit full papers of up to 8 pages of content and short papers of up to 4 pages of content, with unlimited pages for references. Accepted papers will be provided an additional page to address reviewer comments. Previously published papers and those currently undergoing review at other venues are not allowed for submission. Topics of interest include, but are not limited to:

  • Multimodal models and methods for detecting cross domain multimedia contents online, including, but not limited to sentiment analysis, financial analysis, hate speech, cyberbullying, homophobia, threat analysis, disaster prediction, online education, etc.
  • Application of NLP and Computer Vision tools to analyze social media content catered to code mixed low resource languages.
  • Hybrid of NLP, Computer Vision and Speech Processing models for low resource cross-lingual content detection and analysis.
  • - Computational models for multi-modal content detection with emphasis on handling absence of one or more modalities with focus on reuse of foundational models.
  • Development of corpora and annotation guidelines for cross domain, cross modal, multi modal andcross lingual multimedia content analysis
  • Critical evaluation of systems with a focus on Low Resource Cross-Domain, Cross-Lingual and Cross-Modal Multimedia Content Analysis
  • Systems studying model and social biases under cross domain low resource settings for multimedia content analysis.
  • - Cross-domain metrics, which can reliably and robustly measure the quality of system outputs from multiple modalities (e.g., image and speech), different domains (e.g., movie reviews, homophobic contents, sentiment analysis, education content) and different languages with emphasis on foundational models.
  • Study of quality of annotations for cross domain low resource multimedia contents, e.g., consistency of annotations, inter-rater agreement, and bias etc.
  • Applications of foundational models for cross domain low resource multimedia content generation, analysis, and tool creation

Accepted long and short papers will be designated as oral presentations in the workshop, accompanied by publication in SPELLL 2024 conference proceedings