Prompt Equity Transformer (PET)

Product or Programme · All products and programmes
Evidence · 2 claims (2 for) · 1 study (1 causal), q2 · 0 of 1 report an effect size · 2 claims rest on one study

Description

An LLM-based accessibility middleware agent developed by Lier Jin et al. that rewrites heterogeneous user inputs into semantically equivalent, standardized prompts for foundation models.

Components

• Prompt Equity Transformer (PET), an accessibility-oriented prompt normalization agent: PET is an LLM-based agent that automatically transforms heterogeneous user inputs into semantically equivalent prompts with standardized reasoning guidance while preserving the user's original intent. It optimizes expected downstream task performance subject to a predefined semantic-preservation threshold, and does not introduce new information or modify the user's underlying request. Positioned as middleware, it functions as an intelligent accessibility layer between users and foundation models, shifting prompt-optimization responsibility from user to system without modifying the underlying foundation model. (Lier Jin et al. (2026))

Claims

• Applying PET raises accuracy for every user cohort, with the largest gain for the Low Literacy cohort [+M]
• After PET normalization, no statistically significant performance differences remain between user cohorts [+M]

Related Products and Programmes

-

Key Sources

• Lier Jin, Lan Hu, Binqi Shen, Hanyu Cai, and Yuting Xin. (2026). Same Question, Different Answer? Measuring and Mitigating Prompt Privilege for Equitable AI Access. https://arxiv.org/abs/2608.08942

Page metadata

Field	Value
type	product
id	prompt-equity-transformer-pet
title	Prompt Equity Transformer (PET)
description	An LLM-based accessibility middleware agent developed by Lier Jin et al.
status	review
product_kind	software
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.