Pith. sign in

REVIEW 8 cited by

Systematic Biases in LLM Simulations of Debates

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 2402.04049 v3 pith:G3UI7CX2 submitted 2024-02-06 cs.CL cs.AI

classification cs.CLcs.AI
keywords agentsbiasesllmsbehavioralhumansimulationsdebateshumans
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The emergence of Large Language Models (LLMs), has opened exciting possibilities for constructing computational simulations designed to replicate human behavior accurately. Current research suggests that LLM-based agents become increasingly human-like in their performance, sparking interest in using these AI agents as substitutes for human participants in behavioral studies. However, LLMs are complex statistical learners without straightforward deductive rules, making them prone to unexpected behaviors. Hence, it is crucial to study and pinpoint the key behavioral distinctions between humans and LLM-based agents. In this study, we highlight the limitations of LLMs in simulating human interactions, particularly focusing on LLMs' ability to simulate political debates on topics that are important aspects of people's day-to-day lives and decision-making processes. Our findings indicate a tendency for LLM agents to conform to the model's inherent social biases despite being directed to debate from certain political perspectives. This tendency results in behavioral patterns that seem to deviate from well-established social dynamics among humans. We reinforce these observations using an automatic self-fine-tuning method, which enables us to manipulate the biases within the LLM and demonstrate that agents subsequently align with the altered biases. These results underscore the need for further research to develop methods that help agents overcome these biases, a critical step toward creating more realistic simulations.

Discussion (0). Sign in to comment.

Forward citations

Cited by 8 Pith papers

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

  1. More Is Not More: What Matters for Diversity in LLM Opinions?

    cs.CL 2026-05 conditional novelty 7.0 of 10

    Diversity in LLM opinions comes mostly from the first persona sentence and from combining different interaction architectures, not from richer personas, temperature, or diversity instructions.

  2. Point of Order: Action-Aware LLM Persona Modeling for Data-Grounded Civic Deliberation

    cs.CL 2025-11 conditional novelty 6.0 of 10

    Fine-tuning on speaker-attributed, action-tagged transcripts from public meetings lets LLM agents mimic government meeting participants well enough that human judges often cannot tell them from real people.

  3. Revealing Political Bias in LLMs through Structured Multi-Agent Debate

    cs.AI 2025-06 conditional novelty 6.0 of 10

    LLM debate agents with neutral personas lean Democratic, Republican personas drift toward neutral, gender awareness alters stances, and homogeneous groups can show echo chamber attitude intensification.

  4. Generative Exaggeration in LLM Social Agents: Consistency, Bias, and Toxicity

    cs.HC 2025-07 conditional novelty 5.0 of 10

    When LLMs are given more context about a real social media user, they become more ideologically consistent but also more extreme, toxic, and stereotyped than the user actually is.

  5. A Cross-Cultural Comparison of LLM-based Public Opinion Simulation: Evaluating Chinese and U.S. Models on Diverse Societies

    cs.CL 2025-06 conditional novelty 5.0 of 10

    Across U.S. and Chinese survey questions, DeepSeek, GPT-4o, Qwen2.5, and Llama-3.3 all show demographic overgeneralization, with no consistent home-field advantage for the Chinese model.

  6. Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation

    cs.LG 2026-02 conditional novelty 4.0 of 10

    A taxonomy-driven survey arguing that reward design is the central mechanism shaping reliable LLM reasoning, with maps of reward paradigms, reward-hacking failure modes, and benchmark pitfalls.

  7. On the Surprising Efficacy of LLMs for Penetration-Testing

    cs.CR 2025-07 conditional novelty 3.0 of 10

    A critical review arguing that LLMs are surprisingly effective for penetration testing because the task is largely pattern-matching, while noting serious reliability, safety, and cost barriers to autonomous use.

  8. A comprehensive taxonomy of hallucinations in Large Language Models

    cs.CL 2025-08 conditional novelty 2.0 of 10

    A survey that organizes LLM hallucination types, causes, benchmarks, and mitigations, and restates the theorem that hallucination is inevitable for computable LLMs.

Pith tools