{"ID":2850658,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.21448","arxiv_id":"2510.21448","title":"Unified token representations for sequential decision models","abstract":"Transformers have demonstrated strong potential in offline reinforcement learning (RL) by modeling trajectories as sequences of return-to-go, states, and actions. However, existing approaches such as the Decision Transformer(DT) and its variants suffer from redundant tokenization and quadratic attention complexity, limiting their scalability in real-time or resource-constrained settings. To address this, we propose a Unified Token Representation (UTR) that merges return-to-go, state, and action into a single token, substantially reducing sequence length and model complexity. Theoretical analysis shows that UTR leads to a tighter Rademacher complexity bound, suggesting improved generalization. We further develop two variants: UDT and UDC, built upon transformer and gated CNN backbones, respectively. Both achieve comparable or superior performance to state-of-the-art methods with markedly lower computation. These findings demonstrate that UTR generalizes well across architectures and may provide an efficient foundation for scalable control in future large decision models.","short_abstract":"Transformers have demonstrated strong potential in offline reinforcement learning (RL) by modeling trajectories as sequences of return-to-go, states, and actions. However, existing approaches such as the Decision Transformer(DT) and its variants suffer from redundant tokenization and quadratic attention complexity, lim...","url_abs":"https://arxiv.org/abs/2510.21448","url_pdf":"https://arxiv.org/pdf/2510.21448v1","authors":"[\"Zhuojing Tian\",\"Yushu Chen\"]","published":"2025-10-24T13:25:43Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Reinforcement Learning\",\"Transformer\",\"Convolutional Neural Network\"]","has_code":false}