LLMDPD conditions an offline policy-diffusion model on LLM-embedded text task descriptions and a transformer-encoded trajectory prompt, reporting improved success on unseen Meta-World and D4RL tasks, though the evaluation protocol limits the strength of the claim.
Soft adversarial offline reinforcement learning via reducing the attack strength for generalization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
LLM-Driven Policy Diffusion: Enhancing Generalization in Offline Reinforcement Learning
LLMDPD conditions an offline policy-diffusion model on LLM-embedded text task descriptions and a transformer-encoded trajectory prompt, reporting improved success on unseen Meta-World and D4RL tasks, though the evaluation protocol limits the strength of the claim.