Fairness Evaluation and Inference Level Mitigation in LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nadeem, Afrozah, Dras, Mark, Naseem, Usman
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914499880026112
author Nadeem, Afrozah
Dras, Mark
Naseem, Usman
author_facet Nadeem, Afrozah
Dras, Mark
Naseem, Usman
contents Large language models often display undesirable behaviors embedded in their internal representations, undermining fairness, inconsistency drift, amplification of harmful content, and the propagation of unwanted patterns during extended dialogue and conversations. Although training-time or data-centric methods attempt to reduce these effects, they are computationally expensive, irreversible once deployed, and slow to adapt to new conversational contexts. Pruning-based methods provide a flexible and transparent way to reduce bias by adjusting the neurons responsible for certain behaviors. However, most existing approaches are static; once a neuron is removed, the model loses the ability to adapt when the conversation or context changes. To address this, we propose a dynamic, reversible, pruning-based framework that detects context-aware neuron activations and applies adaptive masking to modulate their influence during generation. Our inference-time solution provides fine-grained, memory-aware mitigation with knowledge-preserved, more coherent behavior across multilingual single- and multi-turn dialogues, enabling dynamic fairness control in real-world conversational AI.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18914
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fairness Evaluation and Inference Level Mitigation in LLMs
Nadeem, Afrozah
Dras, Mark
Naseem, Usman
Computation and Language
Artificial Intelligence
Large language models often display undesirable behaviors embedded in their internal representations, undermining fairness, inconsistency drift, amplification of harmful content, and the propagation of unwanted patterns during extended dialogue and conversations. Although training-time or data-centric methods attempt to reduce these effects, they are computationally expensive, irreversible once deployed, and slow to adapt to new conversational contexts. Pruning-based methods provide a flexible and transparent way to reduce bias by adjusting the neurons responsible for certain behaviors. However, most existing approaches are static; once a neuron is removed, the model loses the ability to adapt when the conversation or context changes. To address this, we propose a dynamic, reversible, pruning-based framework that detects context-aware neuron activations and applies adaptive masking to modulate their influence during generation. Our inference-time solution provides fine-grained, memory-aware mitigation with knowledge-preserved, more coherent behavior across multilingual single- and multi-turn dialogues, enabling dynamic fairness control in real-world conversational AI.
title Fairness Evaluation and Inference Level Mitigation in LLMs
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.18914