LLM-Guided Counterfactual Data Generation for Fairer AI
May 13, 2024 · In this paper, we address the data-related aspect of the problem, aiming to enhance the data to guide the model towards greater trustworthiness. Due to their …
In this paper, we present a novel approach for mitigating bias in facial image analysis models. We leverage LLMs and pre trained text-to-image generative models to achieve this.
LLM-Guided Counterfactual Data Generation for Fairer AI
May 13, 2024 · This paper addresses the data-related aspect of the problem, aiming to enhance the data to guide the model towards greater trustworthiness, and observed improvements in …
[2405.04793] Zero-shot LLM-guided Counterfactual Generation: …
May 8, 2024 · We propose a structured pipeline to facilitate this generation, and we hypothesize that the instruction-following and textual understanding capabilities of recent LLMs can be …
We propose a structured pipeline to facilitate this generation, and we hypothesize that the instruction-following and textual understanding capabilities of recent LLMs can be effectively …
LLM-Guided Counterfactual Data Generation for Fairer AI
We analyze the bias metrics and explanations to generate counter-data samples using LLM. - "LLM-Guided Counterfactual Data Generation for Fairer AI" Figure 1: Overview of task-specific …
LLM-Guided Counterfactual Data Generation for Fairer AI
- "LLM-Guided Counterfactual Data Generation for Fairer AI" Figure 2: The figure illustrates the confidence scores before and after fine-tuning the model using counterfactual data for gender …
"LLM-Guided Counterfactual Data Generation for Fairer AI."
May 17, 2024 · Bibliographic details on LLM-Guided Counterfactual Data Generation for Fairer AI.
Prompting Large Language Models for Counterfactual Generation…
May 24, 2023 · Based on this framework, we 1) investigate the strengths and weaknesses of LLMs as the counterfactual generator, and 2) disclose the factors that affect LLMs when …
Accepted Papers | DCAI Workshop - GitHub Pages
LLM-Guided Counterfactual Data Generation for Fairer AI. Dynamic Tiling: A Model-Agnostic, Adaptive, Scalable, and Inference-Data-Centric Approach for Efficient and Accurate Small …
- Some results have been removed