Q-targets make in-context RL less dependent on weak demonstrations
QTPT lets Transformers learn action values from rewards and transitions in their context, instead of mainly copying offline behavior, so in-context reinforcement learning can be more robust when demonstrations are random or suboptimal. The paper replaces behavior cloning with a Bellman-style Q-target objective, proves stronger data-quality robustness in stochastic linear bandits and finite-horizon MDPs, and reports gains on controlled RL benchmarks. It also tests D4RL Kitchen and AntMaze, while noting offline coverage limits remain.