Decision Transformer: Reinforcement Learning via Sequence Modeling

On the reading lists

• #467 in the Top 500 (2026 edition)
• #23 in Reinforcement learning (2026 edition)

Why it matters

Recast reinforcement learning as return-conditioned sequence modeling, linking decision-making with transformers.

Condensed from the expert lists' notes.

The record

First released

2021 (from arXiv 2106.01345)

Citations

2,574 citations, as counted by Papers With Code (CC BY-SA 4.0)

Named by

both expert lists

Score

0.354 · computed rank 467 · consensus 0.13, must-read 0, citations percentile 0.9 (2020s · reinforcement-learning), awards 0, existing lists 0. Method