AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Decision Transformers As Zero-Shot Learners via Text-Behavior Alignment

Xin Zhang, Jonathan Martinez, Yanhua Li, Yingxue Zhang

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Task-inference meta-RL

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add both TBC and TBM alignment objectives (ablation)TG-DT · Ant-dir (zero-shot)test return: 133.6 -> 383.4Improved
add both TBC and TBM alignment objectives (ablation)TG-DT · Cheetah-dir (zero-shot)test return: 859.4 -> 958.4Improved
add text-behavior contrastive loss (TBC) (ablation)TG-DT · Ant-dir (zero-shot)test return: 322.1 -> 383.4Improved
add text-behavior matching loss (TBM) (ablation)TG-DT · Ant-dir (zero-shot)test return: 298.7 -> 383.4Improved
add text-behavior matching loss (TBM) (ablation)TG-DT · Cheetah-dir (zero-shot)test return: 875.2 -> 958.4Improved
share 1 or 2 retrieved related-task trajectories for fine-tuning (K=1 or 2) (ablation)TG-DT · Ant-dir and ML45test returnImproved
increase shared demonstration count K beyond a small value (ablation)TG-DT · data sharingtest returnNo clear effect
text-only zero-shot TG-DT versus DT baselines using test-time interactionTG-DT · Medium datasets zero-shot (5 tasks)test return: Cheetah-dir 549.9 vs 548.9 (PDT)No clear effect
condition on text descriptions instead of language imitation without return conditioning (BC-Z)TG-DT · ML45 zero-shot Mediumtest return: 157.4 -> 309.6Improved
switch from Expert to Mixed dataTG-DT · Cheetah-dir zero-shottest return: 958.4 -> 830.1Hurt
TG-DT versus DPDT on Mixed Cheetah-dirTG-DT · Cheetah-dir Mixed zero-shottest return: 861.2 -> 830.1Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy