# Imprint: Online Memory Compression for Long-Horizon Egocentric QA

arXiv: https://arxiv.org/abs/2607.00696 · HTML: https://mlsanity.com/paper/2607.00696/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 19 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Builds on (most cited first, 8 of 8)

- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [LightRAG: Simple and Fast Retrieval-Augmented Generation](https://mlsanity.com/paper/2410.05779/) (2024)
- [Charades-Ego: A Large-Scale Dataset of Paired Third and First Person Videos](https://mlsanity.com/paper/1804.09626/) (2018)
- [Retrieval-Augmented Generation with Hierarchical Knowledge](https://mlsanity.com/paper/2503.10150/) (2025)
- [The CASTLE 2024 Dataset: Advancing the Art of Multimodal Understanding](https://mlsanity.com/paper/2503.17116/) (2025)
- [Stronger Baselines for Retrieval-Augmented Generation with Long-Context Language Models](https://mlsanity.com/paper/2506.03989/) (2025)

## Abstract (arXiv)

Long-horizon egocentric question answering involves answering about events that have occurred hours or days in the past. This requires memory representations that remain both retrieval-effective and scalable over days or weeks of recording. Existing long-horizon egocentric QA methods construct memory as hierarchical textual summaries of observations. While effective for reducing memory size, summarization optimizes for descriptive compression rather than retrieval: repeated interactions are absorbed into coarse textual descriptions instead of being preserved as explicit, recurring memory units, making long-horizon evidence aggregation difficult. We propose Imprint, an interaction-centric memory framework that formulates long-horizon egocentric memory as an online memory compression problem rather than summarization. Incoming observations are first represented as structured Interaction Records and continuously organized into recurring interaction patterns. Using human memory consolidation signals of recurrence, recency, and distinctiveness, Imprint selectively retains and compresses interactions into a compact retrieval-oriented memory. We evaluate Imprint on EgoLifeQA, a seven-day egocentric benchmark containing questions that require reasoning over interactions occurring hours to days before the query. With the same LLM, Imprint improves QA accuracy from 31.0% to 35.8%, increases evidence-grounded answers by 6× compared with EgoRAG, reduces memory footprint by 2.3×, and decreases retrieval latency by 11.8×. These results demonstrate that memory compression provides a scalable and retrieval-effective foundation for long-horizon egocentric question answering.