Rahaf Aljundi

Researcher@Toyota Motor Europe·比利时·OpenReview

研究方向

Vision Language Models · Privacy leakage · Memory Construction · personalization · Lifelong learning · incremental learning · Domain Adaptation · OOD · Zero shot methods

4.0

Adaptive Vision Encoders: Balancing Efficiency and Robustness in Vision-Language Models

合作者 (20)

Rahaf Aljundi

Online In-Context Distillation for Low-Resource Vision Language Models

OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakage

Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning

Controlling Forgetting with Test-Time Data in Continual Learning

Recurrent Attention-based Token Selection for Efficient Streaming Video-LLMs

Quantum entanglement for attention models

Investigating Grokking phenomena below the Critical Data Regime

Adaptive Vision Encoders: Balancing Efficiency and Robustness in Vision-Language Models