Decision Transformers: the paper trail, organized

What it does This is a curated awesome-list that catalogs research papers on Decision Transformers—the approach that recasts offline reinforcement learning as conditional sequence modeling. It sorts papers by venue and year, tags them with keywords like multi-agent RL, safe RL, and text conditioning, and links out to PDFs and code when available. Think of it as a manually maintained search index for a subfield that has sprawled across arXiv and every major ML conference.

The interesting bit The list does not just dump links; it provides a concise overview of the core DT architecture and advantages (no bootstrapping, no discounting myopia), plus a dedicated taxonomy and survey section to orient newcomers. That context layer is the difference between a folder of PDFs and an actual map.

Key highlights

• Coverage spans offline RL, multi-agent, safe RL, text-conditioned control, and even network intrusion detection.
• Entries include authors, venue, keywords, and experiment environments such as D4RL, Atari, and SMAC.
• Organized chronologically by conference (NeurIPS, ICML, ICLR, etc.) from 2021 through 2025 and beyond.
• Maintained by OpenDILab and actively updated; includes several survey papers for meta-reading.

Verdict Bookmark this if you are writing a literature review or trying to catch a DT variant you half-remember from a poster session; skip it if you are hunting for a unified codebase or framework.