Ruotong Liao, Meikel Kokowski, Xiaoxiao Sun, Yongqin Xian, Laura Bravo Sánchez, Shiye Su, Federico Tombari, Serena Y. Yeung, Volker Tresp · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.30089 · cited by 0 works (OpenAlex) · open-access PDF
Los agentes de coagulación de hoy en día son generalmente invitados a interpretar la evidencia visual presentada a los agentes de coagulación. Para descargar la carga cognitiva de recordar a través de largos horizontes, los asistentes de evaluación de uso deben en cambio adquirir y mantener de forma proactiva la memoria útil a medida que el entorno del usuario evoluciona. Hacer esto en nombre del usuario requiere identificar lo que importa, comprender la escena en tiempo real e iniciar el mantenimiento de la memoria en el fondo mientras el usuario permanece enfocado en sus actividades. Introducimos ProactEgoMem, un benchmark que evalúa la memoria proactiva en el streaming de vídeo egocéntrico. Tres tareas de la evaluación del terreno en las necesidades ambientales: la captura de información (IC) preserva la información proactiva en tiempo real, el estado
Machine translation (Meta M2M100); the English original follows.
Today's multimodal assistants are typically prompted to interpret the visual evidence presented to them. To offload the cognitive burden of remembering across long horizons, wearable assistants should instead proactively acquire and maintain useful memory as the user's surroundings evolve. Doing so on the user's behalf requires identifying what matters, understanding the scene in real time, and initiating memory maintenance in the background while the user stays focused on their activities. We introduce ProactEgoMem, a benchmark that evaluates proactive memory in streaming egocentric video. Three tasks ground the evaluation in ambient environmental needs: Information Capture (IC) preserves transient yet dense visual information, State Tracking (ST) follows changing object states, and Location Tracking (LT) records object movements. The benchmark defines its protocol at two levels of proactiveness: the Notification Level asks for a single timely notification that satisfies the user's intent, while the deeper Intervention Level asks the agent to maintain memory on the user's behalf through an active start-to-stop tool lifecycle. Parallel ProactEgoMem-Daily and ProactEgoMem-Sci tracks contain 300/100 clips covering daily and scientific use cases, paired with 1000/500 questions in ProactEgoMem-QA, which measures the quality of the proactively acquired memory. Across proprietary, open-source, video-language, and streaming baselines, even the strongest models fail to capture memory both timely and accurately. ProactEgoMem provides a rigorous evaluation foundation for proactive multimodal assistance and for video agents that co-exist with their users.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Computer science, Benchmarking, Benchmark (surveying), Human–computer interaction, Protocol (science), Object (grammar)
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30089, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange