6 papers
GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs
Mohamed Eltahir, Lama Ayash, Ali Habibullah +2
Long-video understanding in VLMs is bottlenecked by a single monolithic forward pass over thousands of frames at quadratic attention cost. A common mitigation is to first select a…
VideoAtlas: Navigating Long-Form Video in Logarithmic Compute
Mohamed Eltahir, Ali Habibullah, Yazan Alshoibi +3
Extending language models to video introduces two challenges: representation, where existing methods rely on lossy approximations, and long-context, where caption- or agent-based p…
Vote-in-Context: Turning VLMs into Zero-Shot Rank Fusers
Mohamed Eltahir, Ali Habibullah, Lama Ayash +2
In the retrieval domain, candidates' fusion from heterogeneous retrievers is a long-standing challenge, particularly for complex, multi-modal data such as videos. While typical fus…
AutoArabic: A Three-Stage Framework for Localizing Video-Text Retrieval Benchmarks
Mohamed Eltahir, Osamah Sarraj, Abdulrahman Alfrihidi +4
Video-to-text and text-to-video retrieval are dominated by English benchmarks (e.g. DiDeMo, MSR-VTT) and recent multilingual corpora (e.g. RUDDER), yet Arabic remains underserved,…
Multimodal Lengthy Videos Retrieval Framework and Evaluation Metric
Mohamed Eltahir, Osamah Sarraj, Mohammed Bremoo +5
Precise video retrieval requires multi-modal correlations to handle unseen vocabulary and scenes, becoming more complex for lengthy videos where models must perform effectively wit…
AllWeatherNet:Unified Image Enhancement for Autonomous Driving under Adverse Weather and Lowlight-conditions
Chenghao Qian, Mahdi Rezaei, Saeed Anwar +4
Adverse conditions like snow, rain, nighttime, and fog, pose challenges for autonomous driving perception systems. Existing methods have limited effectiveness in improving essentia…