# Learning Expressive Prompting With Residuals for Vision Transformers

Source: [Learning Expressive Prompting With Residuals for Vision Transformers](<https://www.sota2.com/research/paper/68fa6ee01e209131cea42901>)

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

Data updated: 2025-10-26T01:08:40.604Z

Authors: Rajshekhar Das, Yonatan Dukler, Avinash Ravichandran, Ashwin Swaminathan

Published: 2023-03-27T20:47:01.000Z

## Abstract

Prompt learning is an efficient approach to adapt transformers by inserting learnable set of parameters into the input and intermediate representations of a pre-trained model. In this work, we present Expressive Prompts with Residuals (EXPRES) which modifies the prompt learning paradigm specifically for effective adaptation of vision transformers (ViT). Out method constructs downstream representations via learnable \`\`output'' tokens, that are akin to the learned class tokens of the ViT. Further for better steering of the downstream representation processed by the frozen transformer, we introduce residual learnable tokens that are added to the output of various computations. We apply EXPRES for image classification, few shot learning, and semantic segmentation, and show our method is capable of achieving state of the art prompt tuning on 3/3 categories of the VTAB benchmark. In addition to strong performance, we observe that our approach is an order of magnitude more prompt efficient than existing visual prompting baselines. We analytically show the computational benefits of our approach over weight space adaptation techniques like finetuning. Lastly we systematically corroborate the architectural design of our method via a series of ablation experiments.

## 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.

- [Image Classification on VTAB 1k](<https://www.sota2.com/research/sota/image-classification-on-vtab-1k.md>): Image Classification; VTAB 1K; Overall Mean Accuracy: 72.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2303.15591>)

[PDF](<https://arxiv.org/pdf/2303.15591.pdf>)
