Tag
本文提出 VMem-Bench,一个两阶段诊断基准,系统研究长视频生成中视觉记忆(检索历史画面)对历史一致性与当前故事准确性的权衡影响,发现检索虽提升实体召回率却降低了精确度与禁用实体规避能力。
This paper introduces GEAR, a Geometry-Enabled Attention Routing framework that uses geometry as explicit token-level addresses for visual memory to enhance long-horizon camera-controlled video generation, achieving state-of-the-art results.
KeyRec is a training-free framework for building bounded visual memory in vision-language models, using a recent-observation cache plus a structured event bank to compress long/streaming video context to just 10% of dense visual tokens while setting state-of-the-art results on 13 of 15 streaming and long-video benchmark settings.
This paper proposes VisKG-LM, a method that compiles knowledge graphs into visual memory for efficient multiple-choice question answering, achieving performance gains over baselines by decoupling graph encoding from language reasoning.
A tweet praising CollovLabs' advancement from single-shot recognition to persistent visual memory in AI.
Introduces DMV-Bench, an interactive benchmark for evaluating visual memory in multimodal agents using incidental visual cues from product images, and proposes DualMem, a dual-coding memory architecture that outperforms text-only and other multimodal baselines across various chain lengths.