Pith. sign in

REVIEW 6 cited by

Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2410.02584 v1 pith:4BWUAGOT submitted 2024-10-03 cs.CL cs.CY

classification cs.CLcs.CY
keywords biasesimplicitllmsmitigateinteractionsmulti-agentbiasfine-tuning
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

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.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Breaking Down Bias: On The Limits of Generalizable Pruning Strategies

    cs.CL 2025-02 conditional novelty 6.0 of 10

    Pruning-based bias removal in Llama-3-8B reduces racial bias mainly in the context used to choose what to prune, and transfers poorly across contexts.

  2. Are Human Interactions Replicable by Generative Agents? A Case Study on Pronoun Usage in Hierarchical Interactions

    cs.CL 2025-01 conditional novelty 6.0 of 10

    In a leader/subordinate discussion simulation, most LLM agents do not reproduce the human pronoun pattern, and knowing the pattern does not help them show it.

  3. A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

    cs.CR 2025-04 conditional novelty 5.0 of 10

    A large collaborative survey organizes LLM and LLM-agent safety issues into a full-stack lifecycle framework from data preparation to deployment.

  4. Unmasking Conversational Bias in AI Multiagent Systems

    cs.CL 2025-01 conditional novelty 5.0 of 10

    In simulated echo-chamber chats, conservative-aligned LLM agents often shift to liberal-aligned messages, a drift that one-shot questionnaire tests do not detect.

  5. Bias Mitigation Agent: Optimizing Source Selection for Fair and Balanced Knowledge Retrieval

    cs.AI 2025-08 reject novelty 4.0 of 10

    A multi-agent retrieval system that filters sources by a bias classifier reports an 81.82% relative drop in bias rate, but the evaluation uses the same classifier as the filter.

  6. Multi-agent Systems for Misinformation Lifecycle : Detection, Correction And Source Identification

    cs.MA 2025-05 reject novelty 3.0 of 10

    A conceptual multi-agent architecture for classifying, detecting, correcting, and sourcing misinformation is proposed but not implemented or evaluated.

Pith tools