Multimodal Agent MemorySubmit a paper

Frames2LoRA

Frames2LoRA: Parametric Video Internalization for Vision-Language Models

Manan Suri, Sarvesh Baskar, Dinesh Manocha · 2026-06-04

Read the paperPDF

Management
T1 Static Memory
Memory is built once and reused without change during deployment.
Primary modality
Video & streaming
Also: Image, text
Representation
Internalized latent & parametric

How experience becomes memory

Each path is one memory component, traced from what is stored to how it is found again.

  1. video-conditioned LoRA weights

    1. Represented asExperience-conditioned parametric memoryInternalized latent & parametric
    2. Stored inModular adapter bank
    3. Retrieved byLatent / parametric activation
    4. Source evidenceSource not retained
    A hypernetwork converts video hidden states into LoRA weights used later without visual tokens.Full-paper method / memory representation and retrieval · PDF p. 3 · Figure 1

Related systems