GSAR is a grounding-evaluation framework for multi-agent LLMs that uses a four-way claim typology, evidence-weighted asymmetric scoring, and tiered recovery decisions to detect and mitigate hallucinations.
arXiv preprint arXiv:2411.00784 (2024) 15
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
A reinforcement learning model is ethically fine-tuned using aggregated feedback from LLMs embodying five moral principles via Belief Jensen-Shannon Divergence and Dempster-Shafer Theory.
citing papers explorer
-
GSAR: Typed Grounding for Hallucination Detection and Recovery in Multi-Agent LLMs
GSAR is a grounding-evaluation framework for multi-agent LLMs that uses a four-way claim typology, evidence-weighted asymmetric scoring, and tiered recovery decisions to detect and mitigate hallucinations.
-
Addressing Moral Uncertainty using Large Language Models for Ethical Decision-Making
A reinforcement learning model is ethically fine-tuned using aggregated feedback from LLMs embodying five moral principles via Belief Jensen-Shannon Divergence and Dempster-Shafer Theory.