aiqbench analysis
This research introduces EgoCITE, a framework that enhances long-horizon agentic memory by integrating local multimodal context and temporal awareness into first-person video retrieval systems. By addressing bottlenecks in index reliability and ignoring of time intent, the system achieves significant accuracy gains over baselines while reducing computational costs compared to standard long-context LLM agents. The work advances the state-of-the-art in egocentric question answering through improved indexing schemes and evidence curation tailored for continuous personal experiences.
EgoCITE: Context-Augmented Indexing and Time-Aware Retrieval for Long-Horizon Egocentric Memory
arXiv:2608.12627v2 Announce Type: replace-cross Abstract: Long-horizon egocentric memory transforms continuous first-person video and audio into a searchable record of past experiences. We demonstrate two bottlenecks in existing systems: indices built from context-poor captions are unreliable for agentic search, while retrieval ignores a question's temporal intent. To address both bottlenecks, we introduce EgoCITE (Egocentric Context-augmented Indexing and Time-aware Evidence retrieval), a long-horizon agentic memory framework for egocentric QA. EgoCITE comprises three components. EgoScheme uses local multimodal context to turn fragmentary video captions and speech transcripts into self-contained atomic memory indices. EgoIndex organizes complementary action, activity, utterance, and conversation representations into searchable multi-view memory indices at multiple granularities. EgoRetrv combines semantic search with question-conditioned temporal relevance scoring and curation of retrieved evidence. We evaluate EgoCITE on EgoLifeQA, EgoMem, and EgoR1-Bench in terms of answer accuracy and target-event retrieval alignment. EgoCITE improves accuracy over agentic memory baselines by at least 4.4--14.2% while achieving 36$\times$ lower cost than long-context LLM agents.
Abstract:Long-horizon egocentric memory transforms continuous first-person video and audio into a searchable record of past experiences. We demonstrate two bottlenecks in existing systems: indices built from context-poor captions are unreliable for agentic search, while retrieval ignores a question's temporal intent. To address both bottlenecks, we introduce EgoCITE (Egocentric Context-augmented Indexing and Time-aware Evidence retrieval), a long-horizon agentic memory framework for egocentric QA. EgoCITE comprises three components. EgoScheme uses local multimodal context to turn fragmentary video captions and speech transcripts into self-contained atomic memory indices. EgoIndex organizes complementary action, activity, utterance, and conversation representations into searchable multi-view memory indices at multiple granularities. EgoRetrv combines semantic search with question-conditioned temporal relevance scoring and curation of retrieved evidence. We evaluate EgoCITE on EgoLifeQA, EgoMem, and EgoR1-Bench in terms of answer accuracy and target-event retrieval alignment. EgoCITE improves accuracy over agentic memory baselines by at least 4.4--14.2% while achieving 36$\times$ lower cost than long-context LLM agents.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
arXiv:2608.12627 [cs.CV]
(or arXiv:2608.12627v2 [cs.CV] for this version)
Full text under the publisher's open licence · 176 words
Related coverage
H$^{2}$MT: Semantic Hierarchy-Aware Hierarchical Memory Transformer
arXiv · Aug 19, 2026
Introducing ROCm™ AMD Infinity Context: A Purpose-Built KV Cache Tier for Distributed Inference
AMD · Jul 22, 2026
[AINews] Hot Chips: OpenAI’s Jalapeño, Cerebras CS-5, Groq 3 LPX, Apple M6
Latent Space · Aug 27, 2026
Distributional Approximate Nearest Neighbour Search for Uncertainty-Aware Retrieval
arXiv · Jun 5, 2026