Metraipolis › archive › dc_b2cedb241c92
Language Model Augmented Decision Transformer for Multi-Task Offline Reinforcement Learning — arXiv:2611.04031
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.04031
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T18:38:35.647Z
- SHA-256 of the archived text
b307af48b5752ca60660ce9fd4a3f2362214ef23c4020b722d969820ead241e2- Stable URL of this page
https://metraipolis.com/documento/dc_b2cedb241c92
How to cite
Language Model Augmented Decision Transformer for Multi-Task Offline Reinforcement Learning — arXiv:2611.04031 — arXiv (en), https://arxiv.org/abs/2611.04031 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b2cedb241c92 (Metraipolis archive dc_b2cedb241c92, sha256 b307af48b5752ca60660ce9fd4a3f2362214ef23c4020b722d969820ead241e2)
BibTeX
@misc{dcb2cedb241c92,
title = {Language Model Augmented Decision Transformer for Multi-Task Offline Reinforcement Learning — arXiv:2611.04031},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.04031},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b2cedb241c92; sha256 b307af48b5752ca60660ce9fd4a3f2362214ef23c4020b722d969820ead241e2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04031. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Offline multi-task reinforcement learning (MTRL) aims to learn a unified policy from heterogeneous offline datasets that generalizes across tasks. However, existing offline MTRL methods often struggle with two fundamental issues: severe per-task data scarcity and limited cross-task generalization. In this paper, we propose Language-Model Augmented Decision Transformer (LaMA-DT), a unified sequence-modeling framework that leverages a single large language model (LLM) for both trajectory augmentation and multi-task policy learning. LaMA-DT uses natural-language task descriptions to guide task-aligned trajectory augmentation through a set of supervised generation and alignment objectives, and incorporates a scoring-and-filtering mechanism to retain high-quality augmented trajectories. The same LLM is then repurposed as a language-conditioned Decision Transformer and fine-tuned on a mixture of real and augmented data. Experiments on Meta-World and MuJoCo multi-task benchmarks demonstrate that LaMA-DT consistently outperforms competitive baselines, particularly in low-data regimes, showing that language-guided data augmentation substantially improves cross-task generalization. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.04031 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-04T18:32:52.131Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).