{
  "fig_method_overview": "Paper Banana prompt placeholder: create a clean 16:9 academic system diagram for iMF-AttnRes VLA. Show multi-view images, robot state, and optional language/task conditioning entering a VLA encoder; a policy transformer with AttnRes depth-wise residual aggregation; an improved Mean Flow action generator predicting average flow with one-to-few evaluations; and an exec-length action chunk controlling RoboIMI socket-insert and sim_transfer robots. Use muted conference-paper colors, clear arrows, no decorative 3D, and labels for iMF, AttnRes, infer1/infer2/infer3, and exec16.",
  "fig_socket_reward_latency": "Paper Banana prompt placeholder: create a 4:3 publication-quality reward-latency comparison for socket peg insertion. Plot methods from experimental_log Table 1 with avg_reward on the y-axis and avg_inference_time_ms or inverse latency on the x-axis. Highlight iMF-AttnRes infer2/infer3 at 1472.62/1513.56 avg_reward and 14.409/15.122 ms, diffusion-style infer100 baselines at 861.53/1019.39 avg_reward and 338.291/397.912 ms, ACT at 289.6 and 100.2116 ms, and SmolVLA at 466.16 and 2.614 ms. Use log-scale latency if helpful and label the speed-quality Pareto frontier.",
  "fig_sim_transfer_ablation": "Paper Banana prompt placeholder: create a 4:3 grouped bar chart for sim_transfer ablations from experimental_log Table 2. Show avg_reward and success_like episode count for native diffusion policy, best sim-transfer iMF-AttnRes infer1, ResNet18 multi-token iMF, full-AttnRes vision, and selected horizon/execution variants. Emphasize that best iMF-AttnRes reaches 526.22 avg_reward and 44/100 success-like episodes versus native diffusion policy at 319.2 and 29/100, while several ablations underperform.",
  "fig_imf_attnres_components": "Paper Banana prompt placeholder: create a 16:9 technical diagram contrasting four mathematical components: classical flow matching instantaneous velocity dx_t/dt=v(x_t,t); Mean Flow average velocity over [r,t] with the JVP correction term; improved Mean Flow training relation V_theta(z_t)=u_theta(z_t)+(t-r)JVP_sg(u_theta;v_theta); and AttnRes weighted residual aggregation a_{t+1,s} proportional to exp(w_{t+1} dot RMSNorm(y_s)). Use equation callouts, minimal arrows, and a final callout saying one-to-few-step action generation for robot control."
}