A framework that combines dynamic knowledge-graph mapping, action-space learning, and experience reflection improves LLM agents' scores on interactive fiction games, yet the reported gains are inflated by an unfair baseline comparison and possible training-set leakage.
Hall. To the southwest is the entrance to the Computer Site, and you can go east here as well as go up with a stair
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Learning to Play Like Humans: A Framework for LLM Adaptation in Interactive Fiction Games
A framework that combines dynamic knowledge-graph mapping, action-space learning, and experience reflection improves LLM agents' scores on interactive fiction games, yet the reported gains are inflated by an unfair baseline comparison and possible training-set leakage.