AI Paper Atlas

Research map / Video / Video-language understanding

MemoryFusion: Cross-Temporal Memory Learning for Multimodal Video Fusion

Gong Meiqi, Hao Zhang, Jiayi Ma

NeurIPS 2026 · Video › Video-language understanding · approach: Long video understanding · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes MemoryFusion, a two-stage multimodal video fusion framework using cross-temporal memory learning over historical frames at multiple temporal scales to improve robustness and temporal consistency.

Problem

Multimodal video fusion methods underperform because they insufficiently exploit historical frames and use simplified temporal consistency modeling.

Approach

Builds a two-stage framework on cross-temporal memory learning that leverages historical information across multiple temporal scales and progressively refines temporal consistency.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Temporal Smoothing Module (ablation)MemoryFusion stage 1 · M3SVDflowD: 4.2172 -> 3.6135Improved
add Temporal Smoothing Module (ablation)MemoryFusion stage 1 · M3SVDVIF: 0.4943 -> 0.4877No clear effect
add Short-Term Memory Bank (ablation)MemoryFusion stage 1 · M3SVDVIF: 0.4817 -> 0.4877Improved
add Short-Term Memory Bank (ablation)MemoryFusion stage 1 · M3SVDflowD: 3.6586 -> 3.6135No clear effect
use dual-criterion LTMB update instead of random update (ablation)MemoryFusion stage 1 · M3SVDVIF: 0.4825 -> 0.4877Improved
add Long-Term Memory Bank (ablation)MemoryFusion stage 1 · M3SVDMI: 2.0031 -> 2.1631Improved
add Long-Term Memory Bank (ablation)MemoryFusion stage 1 · M3SVDCC: 0.4646 -> 0.4872Improved
add temporal loss in stage 2 (ablation)MemoryFusion · M3SVDflowD: 3.2065 -> 2.6759Improved
use 3D convolution instead of 2D in stage 2 (ablation)MemoryFusion · M3SVDflowD: 3.5290 -> 2.6759Improved
add stage-2 Temporal Refinement Module (ablation)MemoryFusion · M3SVDflowD: 3.6135 -> 2.6759Improved
apply TRM plug-and-playMMIF-EMMA · M3SVDflowD: 4.6592 -> 2.9549Improved
apply TRM plug-and-playUniVF · M3SVDflowD: 4.0271 -> 2.8025Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy