Current personalization methods for generative vision models typically encode new concepts through continuous adapters or weight updates, yet provide limited control over whether and when a concept should be retrieved. In this work, we introduce Tiny-Engram, a compact trigger-indexed concept table that gives visual memories an explicit lexical address and activation boundary inside frozen image and video generators. Tiny-Engram parameterizes each concept as a small set of memory entries indexed by registered n-gram matches, which modulate text-encoder hidden states only within the matched trigger region. Outside this lexical support, the conditioning pathway is identical to that of the frozen base model. Across both single-encoder latent diffusion and multi-encoder diffusion-transformer backbones, this formulation binds a rare trigger phrase to a target identity while preserving compositional control from the surrounding prompt. We further evaluate the same table-based memory in a text-conditioned video generation setting, where the trigger path reliably alters the generated subject but fine-grained identity persistence across held-out video prompts remains limited. Taken together, these results suggest that small, explicitly addressed concept tables are a practical route to modular visual personalization, with strongest evidence in image generation. For video diffusion, the remaining gap points to a broader requirement: temporally stable identity likely depends on tighter coupling between text-side memory and the evolving visual state, motivating future work on memory injection beyond the text-conditioning interface.
Cite
@article{arxiv.2605.20309,
title = {Tiny-Engram: Trigger-Indexed Concept Tables for Generative Vision},
author = {Runyuan Cai and Yiming Wang and Yu Lin and Xiaodong Zeng},
journal= {arXiv preprint arXiv:2605.20309},
year = {2026}
}