SAGA is a multi-agent LLM framework with scene-graph encoding, tool-augmented per-domain planning, and dual-horizon goal evolution that reports the highest civilization score and infrastructure metrics on FreeCiv while cutting tokens 27%.
EvolveR: Self-evolving LLM agents through an experience-driven lifecycle.arXiv preprint arXiv:2412.04843, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon CivRealm Strategy Planning
SAGA is a multi-agent LLM framework with scene-graph encoding, tool-augmented per-domain planning, and dual-horizon goal evolution that reports the highest civilization score and infrastructure metrics on FreeCiv while cutting tokens 27%.