EM^2Mem: Event-Centric Multimodal Memory for Large Language Models

Hugging Face Daily Papers Papers

Summary

EM^2Mem proposes an event-centric multimodal memory framework that binds heterogeneous evidence to event anchors for compact, generation-ready memory in long-video question answering, improving accuracy and reducing latency.

Multimodal memory offers a scalable interface for long-video question answering, but existing methods often retrieve captions, frames, transcripts, summaries, or graph facts as isolated fragments. Although searchable, such fragments are not generation-ready: language models must reconstruct cross-modal and temporal alignments at inference time, when context is limited and attribution is difficult. We propose EM^2Mem, an event-centric multimodal memory framework that binds heterogeneous evidence to event anchors during memory construction. Each event-indexed memory cell aligns multimodal records, temporal context, graph-linked relations, semantic facts, and provenance, enabling compact evidence readout over grounded multimodal events rather than modality-specific fragments. Across three long-video QA benchmarks, EM^2Mem improves average accuracy over the strongest memory baseline by 2.0, 2.4, and 3.7 points, improves strict event-level Top-5 evidence recall by 7.0 points, and reduces per-query latency by 4.67 times and total inference tokens by 63.66% (The code will be integrated into https://github.com/zjunlp/LightMem).
Original Article
View Cached Full Text

Cached at: 09/02/26, 03:46 AM

Paper page - EM^2Mem: Event-Centric Multimodal Memory for Large Language Models

Source: https://huggingface.co/papers/2609.00551 Authors:

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

EM²Mem binds multimodal evidence to event anchors for compact, generation-ready memory in long-video question answering.

Multimodal memoryoffers a scalable interface for long-video question answering, but existing methods often retrieve captions, frames, transcripts, summaries, or graph facts as isolated fragments. Although searchable, such fragments are not generation-ready: language models must reconstruct cross-modal and temporal alignments at inference time, when context is limited and attribution is difficult. We propose EM^2Mem, an event-centricmultimodal memoryframework that binds heterogeneous evidence to event anchors during memory construction. Each event-indexed memory cell aligns multimodal records,temporal context,graph-linked relations, semantic facts, andprovenance, enabling compact evidence readout over grounded multimodal events rather than modality-specific fragments. Across threelong-video QAbenchmarks, EM^2Mem improves average accuracy over the strongest memory baseline by 2.0, 2.4, and 3.7 points, improves strict event-level Top-5evidence recallby 7.0 points, and reduces per-query latency by 4.67 times and total inference tokens by 63.66% (The code will be integrated into https://github.com/zjunlp/LightMem).

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2609\.00551

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.00551 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.00551 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.00551 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Parametric Multimodal User Memory: Storing What Captions Cannot Carry

arXiv cs.CL

This research paper introduces a parametric multimodal user memory system that improves AI agents' ability to recall users by integrating perceptual data like voice and appearance, using vision-language models and dedicated encoders to surpass text-based methods.

δ-mem: Efficient Online Memory for Large Language Models

Hugging Face Daily Papers

The paper introduces δ-mem, a lightweight memory mechanism that enhances large language models by augmenting a frozen attention backbone with a compact associative memory state. It demonstrates improved performance on memory-heavy benchmarks with minimal computational overhead.