FEP-Diff uses a dual-branch spatiotemporal encoder, goal-conditioned belief learner optimized by free-energy objective with social consistency constraint, and residual diffusion generator to outperform prior methods on five benchmarks under restricted observability.
IEEE Transactions on Intelligent Vehicles (2024), 1–33
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
-
Agent-Centric Social Trajectory Prediction: A Free Energy Principle Perspective
FEP-Diff uses a dual-branch spatiotemporal encoder, goal-conditioned belief learner optimized by free-energy objective with social consistency constraint, and residual diffusion generator to outperform prior methods on five benchmarks under restricted observability.