Hallucination Mitigation using Agentic AI Natural Language-Based Frameworks
cs.CL, cs.AI, cs.MA
Submitted: 2025-01-19
Updated: 2026-11-16
Comments: v2: 19 pages, 6 figures, corrected the sign of FGR in the Total Hallucination Score and recomputed all results and figures from the released per-prompt data; relative (percentage) changes replaced with absolute changes and effect sizes; corrected the description of the agent orchestration; OVON renamed Open Floor Protocol (OFP)
Code: https://github.com/open-voice-interoperability/docs
Project page: https://microsoft.github.io/autogen
License: http://creativecommons.org/licenses/by/4.0/
The gist: Hallucinations remain a significant challenge for Generative AI, undermining trust in AI systems and their reliability.
Terminology
Abstract
Hallucinations remain a significant challenge for Generative AI, undermining trust in AI systems and their reliability. This study investigates how orchestrating multiple specialized AI agents can mitigate hallucinations, with a focus on systems that use Natural Language Processing (NLP) for agent interaction. We design a pipeline that feeds 310 prompts, purposefully crafted to induce hallucinations, into a front-end agent; its outputs are reviewed and refined by second- and third-level agents that detect unverified claims, add explicit disclaimers and clarify speculative content. We introduce four novel Key Performance Indicators (KPIs) - Factual Claim Density, Factual Grounding References, Fictional Disclaimer Frequency and Explicit Contextualization Score - aggregated into a Total Hallucination Score (THS) and assessed by a dedicated fourth-level evaluator agent. Inter-agent context is carried by the Open Floor Protocol (OFP, formerly OVON, Open Voice Network), whose structured JSON messages let each reviewer pass its assessment of hallucination likelihood, and the reasons behind it, to the next stage. Across the 310 prompts, THS improves by a mean of-0.131 per prompt from the front-end agent to the final reviewer (Cohen's d = 2.38), with lower scores on 302 of 310 prompts. The gain is carried mainly by explicit disclaimers and contextualization, making speculative content more transparent and clearly demarcated from factual claims. The results show that multi-agent orchestration with structured, NLP-based exchange of meta-information can substantially strengthen the epistemic framing of AI-generated responses and improve their explainability.
Sources
- Agent AI: Surveying the Horizons of Multimodal Interaction
- Conversational AI Multi-Agent Interoperability, Universal Open APIs for Agentic Natural Language Multimodal Communications
- AI Multi-Agent Interoperability Extension for Managing Multiparty Conversations
- Good Parenting is all you need -- Multi-agentic LLM Hallucination Mitigation
- Interpreting and Mitigating Hallucination in MLLMs through Multi-agent Debate
- Measuring Attribution in Natural Language Generation Models
- Hallucination is Inevitable: An Innate Limitation of Large Language Models
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data