A multi-agent LLM debate with four semantic-relation experts builds causal event graphs that improve explainable event likelihood prediction and match fine-tuned models on forecasting and next-event prediction.
The gap is very small for the causality whereas the longer chains lead to larger gap between the CGEL and the baseline in terms of informativeness and coherence
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Causal Graph based Event Reasoning using Semantic Relation Experts
A multi-agent LLM debate with four semantic-relation experts builds causal event graphs that improve explainable event likelihood prediction and match fine-tuned models on forecasting and next-event prediction.