Détail du document
Identifiant
oai:arXiv.org:2410.02584
Sujet
Computer Science - Computation and... Computer Science - Computers and S...Auteur
Borah, Angana Mihalcea, RadaCatégorie
Computer Science
Année
2024
Date de référencement
09/10/2024
Mots clés
multi-agent interactions implicitRésumé
As Large Language Models (LLMs) continue to evolve, they are increasingly being employed in numerous studies to simulate societies and execute diverse social tasks.
However, LLMs are susceptible to societal biases due to their exposure to human-generated data.
Given that LLMs are being used to gain insights into various societal aspects, it is essential to mitigate these biases.
To that end, our study investigates the presence of implicit gender biases in multi-agent LLM interactions and proposes two strategies to mitigate these biases.
We begin by creating a dataset of scenarios where implicit gender biases might arise, and subsequently develop a metric to assess the presence of biases.
Our empirical analysis reveals that LLMs generate outputs characterized by strong implicit bias associations (>= 50\% of the time).
Furthermore, these biases tend to escalate following multi-agent interactions.
To mitigate them, we propose two strategies: self-reflection with in-context examples (ICE); and supervised fine-tuning.
Our research demonstrates that both methods effectively mitigate implicit biases, with the ensemble of fine-tuning and self-reflection proving to be the most successful.
;Comment: Accepted to EMNLP Findings 2024
Borah, Angana,Mihalcea, Rada, 2024, Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions
Systematic Feature Design for Cycle Life Prediction of Lithium-Ion Batteries During Formation
cycle prediction framework feature design features life formation
