Post-training pipeline that boosts game-control transfer by fine-tuning on 6,044 curated gameplay trajectories and then continuing training with PPO across 16 synthetic 2D/3D environments, extending the keyboard/mouse action space with relative pointer motion, sustained inputs, and composite actions.
Answers spatial choice, scalar, and point questions by classifying among discretized candidates using pretrained attention projections from a Qwen3.5-4B backbone instead of autoregressive text generation. Trains on converted spatial QA data, adopts coarse-to-fine refinement and joint multi-stage supervision, and reports higher accuracy with several-fold faster inference than autoregressive baselines.