Safety
ASPECT:Analogical Semantic Policy Execution via Language Conditioned Transfer
arXiv:2604.08355v2 Announce Type: replace Abstract: Reinforcement Learning (RL) agents often struggle to generalize knowledge to new tasks, even those structurally similar to ones they have mastered.
arXiv:2604.08355v2 Announce Type: replace Abstract: Reinforcement Learning (RL) agents often struggle to generalize knowledge to new tasks, even those structurally similar to ones they have mastered. Although recent approaches have attempted to mitigate this issue via zero-shot transfer, they are often constrained by predefined, discrete class systems, limiting their adaptability to novel or compositional task variations. We propose a significantly more generalized approach, replacing discrete latent variables with natural language conditioning via a text-conditioned Variational Autoencoder (VAE). Our core innovation utilizes a Large Language Model (LLM) as a dynamic extit{semantic operator} at test time. Rather than relying on rigid rules, our agent queries the LLM to semantically remap the description of the current observation to align with the source task. This source-aligned caption conditions the VAE to generate an imagined state compatible with the agent's original training, enabling direct policy reuse. By harnessing the flexible reasoning capabilities of LLMs, our approach achieves zero-shot transfer across a broad spectrum of complex and truly novel analogous tasks, moving beyond the limitations of fixed category mappings. Code and videos are available href{https://anonymous.4open.science/r/ASPECT-85C3/}{here}.
Related
- Reason in Chains, Learn in Trees: Self-Rectification and Grafting for Multi-turn Agent Policy Optimization
- Dejavu: Towards Experience Feedback Learning for Embodied Intelligence
- ChipSeek: Optimizing Verilog Generation via EDA-Integrated Reinforcement Learning
Source: arXiv cs.AI | 2026-04-13