# ELT: Elastic Looped Transformers for Visual Generation

Source: [ELT: Elastic Looped Transformers for Visual Generation](<https://www.sota2.com/research/paper/69dc786c7d0172140ec789af>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-25T05:00:05.609Z

Authors: Sahil Goyal, Swayam Agrawal, Gautham Govind Anil, Prateek Jain, Sujoy Paul, Aditya Kusupati

Published: 2026-04-10T09:53:27.000Z

## Abstract

We introduce Elastic Looped Transformers (ELT), a highly parameter-efficient class of visual generative models based on a recurrent transformer architecture. While conventional generative models rely on deep stacks of unique transformer layers, our approach employs iterative, weight-shared transformer blocks to drastically reduce parameter counts while maintaining high synthesis quality. To effectively train these models for image and video generation, we propose the idea of Intra-Loop Self Distillation (ILSD), where student configurations (intermediate loops) are distilled from the teacher configuration (maximum training loops) to ensure consistency across the model's depth in a single training step. Our framework yields a family of elastic models from a single training run, enabling Any-Time inference capability with dynamic trade-offs between computational cost and generation quality, with the same parameter count. ELT significantly shifts the efficiency frontier for visual synthesis. With $4\\times$ reduction in parameter count under iso-inference-compute settings, ELT achieves a competitive FID of $2.0$ on class-conditional ImageNet $256 \\times 256$ and FVD of $72.8$ on class-conditional UCF-101.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Video Generation on UCF-101](<https://www.sota2.com/research/sota/video-generation-on-ucf-101.md>): Video Generation; UCF-101; FVD: 60.8.

- [Class-conditional Image Generation on ImageNet 256x256](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256.md>): Class-conditional Image Generation; ImageNet 256x256; Inception Score (IS): 281.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.09168>)

[PDF](<https://arxiv.org/pdf/2604.09168.pdf>)
