Where Is the Truth When AI Agents Disagree? Dependence-Aware Belief Propagation for Unsupervised Truth Discovery (Extended Version)
cs.AI, cs.CL, cs.DB, stat.ML
Submitted: 2026-11-03
Updated: 2026-11-03
Code: https://github.com/LaureBerti/BPTD
Terminology
Sources
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Mixture-of-Agents Enhances Large Language Model Capabilities
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- Language Models (Mostly) Know What They Know
- Correlated Errors in Large Language Models
- Rethinking Mixture-of-Agents: Is Mixing Different Large Language Models Beneficial?
- Nine Judges, Two Effective Votes: Correlated Errors Undermine LLM Evaluation Panels
- Truth Discovery Algorithms: An Experimental Evaluation
- More Agents Is All You Need
- Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
- Beyond Majority Voting: LLM Aggregation by Leveraging Higher-Order Information
- Dependence-Aware Label Aggregation for LLM-as-a-Judge via Ising Models
- CARE: Confounder-Aware Aggregation for Reliable LLM Evaluation
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents