MediaEval 2026
Multimedia Benchmark Workshop

Table of Contents

Memorability

Overview paper:

• Overview of The MediaEval 2026 Predicting Movie and Commercial Memorability Task
  Iván Martín-Fernández, Aashutosh Ganesh, Mihai Gabriel Constantin, Claire-Hélène Demarty, Manuel Gil-Martín, Sebastian Halder, Bogdan Ionescu, Rukiye Savran Kiziltepe, Ana Matran-Fernandez, Alba G. Seco de Herrera

Working notes papers:

• Multimodal Learning Framework for Commercial Video Memorability Prediction
  Mahnoor Adeel, Kisa Fatima, Muhammad Ibrahim Ayoubi, Muhammad Atif Tahir, Mustafa Usmani

• A Study on Multi-modal Ensemble Models for Commercial Video Memorability Prediction
  Olufela Fieulleteau, Conner Perriello, Sejong Yoon

• A Hand-Crafted EEG Feature Bank with Stability-Based Selection for Movie Recall Detection
  Hao-Tien Yu, Yi-En Dong

• Predicting Long-Term Movie Recall from EEG with Deep Models
  Iván Martín-Fernández, Jaime León, Sergio Esteban-Romero, Manuel Gil-Martín, Fernando Fernández-Martínez

• Visual Transformers and Contextual Features for Movie Memorability Estimation
  Aashutosh Ganesh, Mirela Popa, Nava Tintarev

• Decoding Memory at the Neural Level: wPLI Connectivity and Quantum Tensor Networks for EEG-Based Memorability Prediction
  Berkay Bayramoğlu, İsmail Erol, Sohel Akhtar, Rukiye Savran Kiziltepe, Murat Karakus

• A CLIP-Based Visual-Semantic Embedding Approach for Commercial Memorability Prediction
  Elif Nur Tekay

Quest for insight papers:

• Do Better-Looking Videos Stay in Memory? A VBench Study of Commercial and Brand Memorability
  Hao-Tien Yu, Yi-En Dong

• Quest for Insight: Contextual and Popularity Factors in Movie Memorability Prediction
  Aashutosh Ganesh, Iván Martín-Fernández, Mirela Popa, Manuel Gil-Martín, Fernando Fernández-Martínez, Nava Tintarev

NewsImages

Overview paper:

• NewsImages in MediaEval 2026: Automated Image Recommendations with Retrieval and Generation Techniques for News Articles Thumbnails
  Lucien Heitz, Bruno N. Sotic, Ali A. Katamjani, Qi Bi, Bram Bakker, Luca Rossetto, Jaap Kamps

Working notes papers:

• A Test-time Actor-Critic Approach to News Images Generation
  Damianos Galanopoulos, Vasileios Mezaris

• Hybrid Retrieval and Generative Image Recommendation for News Articles: The FAST-MS(DS) Approach at MediaEval 2026
  Aqsa Khan Jadoon, Muhammad Rafi

• Illustration Baseline for the NewsImages 2026 Challenge
  Bruno Nadalic Sotic, Jaap Kamps

• MediaEval 2026 Grand Challenge: NewsImages
  Chi Chong Wong, Sam Ziaie Kondori

• Generating Images from News Titles via Structured Prompt Generation
  Runze Li

Quest for insight papers:

• Assessing the Viability of Image-to-Image Models for Restoring Historical Photographs of News Articles
  Lucien Heitz, Bruno N. Sotic, Qi Bi, Jaap Kamps

• Operationalising Journalistic and Indexical Dimensions of AI Imagery
  Bruno Nadalic Sotic, Jaap Kamps

Synthetic Images

Overview paper:

• Synthetic Images at MediaEval 2026: Advancing detection of generative AI in real-world online images
  Olga Papadopoulou, Dimitrios Karageorgiou, Christos Koutlis, Efstratios Gavves, Hannes Mareen, Symeon Papadopoulos

Working notes papers:

• Layer-Wise CLIP Feature Fusion for Synthetic Image Detection
  Jeng Wen Joshua Lean

• Multi-Backbone Frozen Features and OOF-Stacked Ensembling for Synthetic Image Detection
  Syed Saad Hasan Emad, Abdulrehman Muhammad, Hammad Shariq, Syed Faran Tariq, Syed Muhammad Meesum Abbas, Syed Daniyal Hussain, Muhammad Atif Tahir

• VLM Fine-tuning and Multi-Model Ensemble for Synthetic Images Detection
  Hengwei Zhao, Hori Takayuki

• Detecting Synthetic Images with Frozen DINOv3 Layer Features: A Layer-Depth Study
  Vedant Gupta

• Exploring Swin Transformers, Luminance Input, and Hybrid Architectures for Synthetic Image Detection
  Amrit Gopinath, Raghul, Mirunalini P

• A Progressive Pipeline from CNNs to CLIP-Based Feature Extraction for Synthetic Image Detection at MediaEval 2026
  Maham Junaid, Maryam Ihsan, Hamna Usman, Atif Tahir

• Synthetic Image Detection under Distribution Shift: CLIP-FFT Fusion and ForensicGAT
  Berkay Bayramoğlu, İsmail Erol, Murat Karakus, Rukiye Savran Kiziltepe

Missing Pieces

Overview paper:

• Missing Pieces and Misinformation: Shared Task on Detecting and Generating Implicit Components in Enthymematic Arguments
  Martial Pastor, Nelleke Oostdijk

Working notes papers:

• Learning from Annotator Disagreement for Enthymeme Detection and Reconstruction on Social Media
  Moez Ur Rehman, Haiyan Alam

• Baseline Methods for the Missing Pieces and Misinformation Task of MediaEval 2026
  Martial Pastor, Stefan Kok

Quest for insight papers:

• Does Annotator Disagreement Signal Enthymeme Presence? Leveraging Inter-Annotator Uncertainty for Implicit Proposition Detection
  Afifah Luqman Berry, Hadiya Ebrahim

• Does Modelling Annotator Distributions Help? Evidence and a Metric Caveat from Enthymeme Detection
  Moez Ur Rehman, Haiyan Alam

• In-Domain Encoders Beat Scale: A Stable, Honestly-Calibrated System for Enthymeme Detection and Reconstruction
  Muhammad Rafi, Ajay Kumar, Abdul Rehman Azam

• Enthymeme Detection and Implicit Proposition Generation: An Ensemble Approach Leveraging Annotator Disagreement
  Mariya Joevita, Priya Verma

Medico

Overview paper:

• Medico 2026: Visual Question Answering for Gastrointestinal Imaging
  Sushant Gautam, Vajira Thambawita, Michael Riegler, Pål Halvorsen, Steven Hicks

Working notes papers:

• CARES at MediaEval Medico 2026: Calibrated Aspect-Reliable Explanations and Safety for Gastrointestinal Visual Question Answering
  Syed Saad Hasan Emad, Itbaan Safwan, Muhammad Atif Tahir

• Boundary-Guided Multimodal Reasoning for Explainable Gastrointestinal VQA
  Tien-Dat Dam, Le-Tran Nguyen, Trung-Nghia Le

• Do Medical Vision Encoders Help? PEFT Benchmarking of Vision-Language Models for Gastrointestinal VQA
  Matheus Mattioli, Jurandy Almeida

• CATA: Clinical-Aware Topological Adaptation for Generative Medical VQA
  Quang Nguyen, Binh T. Nguyen

MultiSumm

Overview paper:

• MultiSumm - Multimodal Summarisation Task at MediaEval 2026
  Anastasiia Potiagalova, Hyunji Cho, Gareth Jones

Working notes papers:

• MultiSummGRAG: A Locally-Hosted GraphRAG Pipeline for the MultiSumm Task at MediaEval 2026
  Kimia Rabishokr, Andreas Weiler