Call For Papers

MMLow 2026 invites submissions on, but not limited to, the following topics, with a strong emphasis on academic diversity, inclusivity, and low-resource settings.

• Low-Resource and Inclusive Multimodal AI

  • Multilingual and multimodal machine learning methods for low-resource and underrepresented languages

  • Indigenous, tribal, and endangered language technologies, including dataset creation and evaluation

  • Multimodal datasets and benchmarks across diverse scripts, modalities, and cultural contexts

  • Scalable and transferable methods that generalize across languages, domains, and modalities

  • Bias, fairness, and inclusivity in multimodal NLP systems

  • Content safety and moderation in multilingual and multimodal settings

  • Language technologies addressing gender diversity and LGBTQ+ communities

  • Culturally grounded NLP and multimodal AI, including region-specific and community-driven approaches

  • Evaluation frameworks and metrics for inclusive and low-resource multimodal NLP

• Multimodal Safety, Hate Speech, and Social Media

  • Multimodal sentiment analysis in regional languages

  • Hate content video detection in regional languages

  • Trolling, offensive language, and cyberbullying detection in memes and social media discourse

  • Multimodal data fusion and representation for hate speech detection in regional languages

  • Benchmark datasets and evaluation for multimodal hate speech in regional languages

• Core Multimodal Technologies

  • Multimodal embeddings and architectures

  • Multimodal dataset creation and annotation

  • Feature extraction and feature fusion techniques

  • Transfer learning and cross-modality adaptation

  • Generative models for multimodal data augmentation

  • Advanced prompting techniques for multimodal inference in low-resource languages

  • Handling hallucination and improving context-aware multimodal generation

• Retrieval, Generation, and Applications

  • Text-to-image retrieval and generation

  • Text-to-video generation

  • Text-to-sign language and video-based sign language translation

  • Multimodal recommender systems

  • Multimodal question answering and summarization in low-resource languages

  • Multimodal content generation

• Multimodal Pretraining, Translation, and Resource Efficiency

  • Pretraining and fine-tuning multimodal models for low-resource contexts

  • Resource-efficient and scalable multimodal systems

  • Multimodal machine translation for low-resource languages

  • Speech-to-text and speech-to-speech multimodal systems

• Evaluation and Preservation

  • Evaluation metrics for multimodal systems in low-resource settings

  • Multimodal approaches for endangered language preservation

  • Visual and audio-text alignment for indigenous or heritage language resources