DAIL: Beyond Task Ambiguity for Language-Conditioned Reinforcement Learning
Abstract
Comprehending natural language and following human instructions are critical capabilities for intelligent agents. However, the flexibility of linguistic instructions induces substantial ambiguity across language-conditioned tasks, severely degrading algorithmic performance. To address these limitations, we present a novel method named DAIL (Distributional Aligned Learning), featuring two key components: distributional policy and semantic alignment. Specifically, we provide theoretical results that the value distribution estimation mechanism enhances task differentiability. Meanwhile, the semantic alignment module captures the correspondence between trajectories and linguistic instructions. Extensive experimental results on both structured and visual observation benchmarks demonstrate that DAIL effectively resolves instruction ambiguities, achieving superior performance to baseline methods. Our implementation is available at https://github.com/RunpengXie/Distributional-Aligned-Learning.
Cite
@article{arxiv.2510.19562,
title = {DAIL: Beyond Task Ambiguity for Language-Conditioned Reinforcement Learning},
author = {Runpeng Xie and Quanwei Wang and Hao Hu and Zherui Zhou and Ni Mu and Xiyun Li and Yiqin Yang and Shuang Xu and Qianchuan Zhao and Bo XU},
journal= {arXiv preprint arXiv:2510.19562},
year = {2025}
}
Comments
Website at: https://github.com/RunpengXie/Distributional-Aligned-Learning