@inproceedings{kwon-etal-2026-forgetting,
    title = "Before Forgetting, Learn to Remember: Revisiting Foundational Learning Failures in {LVLM} Unlearning Benchmarks",
    author = "Kwon, Junehyoung  and
      Kim, MiHyeon  and
      Lee, Eunju  and
      Yun, JungMin  and
      Lim, Byeonggeuk  and
      Kim, YoungBin",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1701/",
    pages = "34063--34079",
    ISBN = "979-8-89176-395-1",
    abstract = "While Large Vision-Language Models (LVLMs) offer powerful capabilities, they pose privacy risks by unintentionally memorizing sensitive personal information. Current unlearning benchmarks attempt to mitigate this using fictitious identities but overlook a critical \textit{stage 1 failure}: models fail to effectively memorize target information initially, rendering subsequent unlearning evaluations unreliable. Diagnosing under-memorization and the multi-hop curse as root causes, we introduce ReMem, a Reliable Multi-hop and Multi-image Memorization Benchmark. ReMem ensures robust foundational learning through principled data scaling, reasoning-aware QA pairs, and diverse visual contexts. Additionally, we propose a novel Exposure metric to quantify the depth of information erasure from the model{'}s internal probability distribution. Extensive experiments demonstrate that ReMem provides a rigorous and trustworthy framework for diagnosing both learning and unlearning behaviors in LVLMs."
}