Update README.md

README.md

@@ -53,6 +53,1332 @@ model-index:
           - type: iqm_normalized_95
             value: 0.99
             name: Normalized Score IQM (95% CI)
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: ant_v4
+        metrics:
+          - type: total_reward
+            value: 6315.00 +/- 675.00
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.10
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: halfcheetah_v4
+        metrics:
+          - type: total_reward
+            value: 7226.50 +/- 241.50
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.93 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: hopper_v4
+        metrics:
+          - type: total_reward
+            value: 2794.60 +/- 612.62
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.86 +/- 0.19
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: humanoid_v4
+        metrics:
+          - type: total_reward
+            value: 7376.26 +/- 0.00
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.00
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: humanoidstandup_v4
+        metrics:
+          - type: total_reward
+            value: 320567.82 +/- 58462.11
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.02 +/- 0.21
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: inverteddoublependulum_v4
+        metrics:
+          - type: total_reward
+            value: 6105.75 +/- 4368.65
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.65 +/- 0.47
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: invertedpendulum_v4
+        metrics:
+          - type: total_reward
+            value: 1000.00 +/- 0.00
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.00
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: pusher_v4
+        metrics:
+          - type: total_reward
+            value: -37.82 +/- 8.72
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.02 +/- 0.08
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: reacher_v4
+        metrics:
+          - type: total_reward
+            value: -6.25 +/- 2.63
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.07
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: swimmer_v4
+        metrics:
+          - type: total_reward
+            value: 93.20 +/- 5.40
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.06
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: MuJoCo
+          type: walker2d_v4
+        metrics:
+          - type: total_reward
+            value: 5400.00 +/- 107.95
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: assembly-v2
+        metrics:
+          - type: total_reward
+            value: 307.08 +/- 25.20
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.04 +/- 0.10
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: basketball-v2
+        metrics:
+          - type: total_reward
+            value: 568.04 +/- 60.72
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.02 +/- 0.11
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: bin-picking-v2
+        metrics:
+          - type: total_reward
+            value: 7.88 +/- 4.28
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.01 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: box-close-v2
+        metrics:
+          - type: total_reward
+            value: 61.75 +/- 13.54
+            name: Total reward
+          - type: normalized_total_reward
+            value: -0.04 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: button-press-v2
+        metrics:
+          - type: total_reward
+            value: 624.67 +/- 42.77
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.07
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: button-press-topdown-v2
+        metrics:
+          - type: total_reward
+            value: 449.36 +/- 62.16
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.94 +/- 0.14
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: button-press-topdown-wall-v2
+        metrics:
+          - type: total_reward
+            value: 482.08 +/- 32.48
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.07
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: button-press-wall-v2
+        metrics:
+          - type: total_reward
+            value: 672.00 +/- 26.48
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.04
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: coffee-button-v2
+        metrics:
+          - type: total_reward
+            value: 719.00 +/- 41.10
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.06
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: coffee-pull-v2
+        metrics:
+          - type: total_reward
+            value: 26.04 +/- 56.12
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.07 +/- 0.20
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: coffee-push-v2
+        metrics:
+          - type: total_reward
+            value: 571.01 +/- 112.28
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.01 +/- 0.20
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: dial-turn-v2
+        metrics:
+          - type: total_reward
+            value: 783.90 +/- 53.17
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.07
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: disassemble-v2
+        metrics:
+          - type: total_reward
+            value: 523.60 +/- 58.15
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.12
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: door-close-v2
+        metrics:
+          - type: total_reward
+            value: 538.10 +/- 25.76
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.02 +/- 0.05
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: door-lock-v2
+        metrics:
+          - type: total_reward
+            value: 356.51 +/- 249.44
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.35 +/- 0.36
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: door-open-v2
+        metrics:
+          - type: total_reward
+            value: 581.33 +/- 26.33
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.05
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: door-unlock-v2
+        metrics:
+          - type: total_reward
+            value: 352.86 +/- 147.78
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.21 +/- 0.26
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: drawer-close-v2
+        metrics:
+          - type: total_reward
+            value: 838.88 +/- 7.41
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.96 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: drawer-open-v2
+        metrics:
+          - type: total_reward
+            value: 493.00 +/- 3.57
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: faucet-close-v2
+        metrics:
+          - type: total_reward
+            value: 749.46 +/- 14.83
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: faucet-open-v2
+        metrics:
+          - type: total_reward
+            value: 732.47 +/- 15.23
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: hammer-v2
+        metrics:
+          - type: total_reward
+            value: 669.31 +/- 69.56
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.12
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: hand-insert-v2
+        metrics:
+          - type: total_reward
+            value: 142.81 +/- 146.64
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.19 +/- 0.20
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: handle-press-v2
+        metrics:
+          - type: total_reward
+            value: 835.30 +/- 114.19
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.15
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: handle-press-side-v2
+        metrics:
+          - type: total_reward
+            value: 852.96 +/- 16.08
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: handle-pull-v2
+        metrics:
+          - type: total_reward
+            value: 701.10 +/- 13.82
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: handle-pull-side-v2
+        metrics:
+          - type: total_reward
+            value: 493.10 +/- 53.65
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.11
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: lever-pull-v2
+        metrics:
+          - type: total_reward
+            value: 548.72 +/- 81.12
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.96 +/- 0.16
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: peg-insert-side-v2
+        metrics:
+          - type: total_reward
+            value: 352.43 +/- 137.24
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.01 +/- 0.40
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: peg-unplug-side-v2
+        metrics:
+          - type: total_reward
+            value: 401.52 +/- 175.27
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.75 +/- 0.34
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: pick-out-of-hole-v2
+        metrics:
+          - type: total_reward
+            value: 364.20 +/- 79.56
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.91 +/- 0.20
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: pick-place-v2
+        metrics:
+          - type: total_reward
+            value: 414.02 +/- 91.10
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.22
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: pick-place-wall-v2
+        metrics:
+          - type: total_reward
+            value: 553.18 +/- 84.72
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.04 +/- 0.16
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: plate-slide-v2
+        metrics:
+          - type: total_reward
+            value: 531.98 +/- 156.94
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.34
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: plate-slide-back-v2
+        metrics:
+          - type: total_reward
+            value: 703.93 +/- 108.27
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.16
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: plate-slide-back-side-v2
+        metrics:
+          - type: total_reward
+            value: 721.29 +/- 62.15
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.09
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: plate-slide-side-v2
+        metrics:
+          - type: total_reward
+            value: 578.24 +/- 143.73
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.83 +/- 0.22
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: push-v2
+        metrics:
+          - type: total_reward
+            value: 729.33 +/- 104.40
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.14
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: push-back-v2
+        metrics:
+          - type: total_reward
+            value: 372.16 +/- 112.75
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.29
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: push-wall-v2
+        metrics:
+          - type: total_reward
+            value: 741.68 +/- 14.84
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: reach-v2
+        metrics:
+          - type: total_reward
+            value: 684.45 +/- 136.55
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.01 +/- 0.26
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: reach-wall-v2
+        metrics:
+          - type: total_reward
+            value: 738.02 +/- 100.96
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.17
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: shelf-place-v2
+        metrics:
+          - type: total_reward
+            value: 268.34 +/- 29.07
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.01 +/- 0.11
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: soccer-v2
+        metrics:
+          - type: total_reward
+            value: 438.44 +/- 189.63
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.80 +/- 0.35
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: stick-pull-v2
+        metrics:
+          - type: total_reward
+            value: 483.98 +/- 83.25
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.92 +/- 0.16
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: stick-push-v2
+        metrics:
+          - type: total_reward
+            value: 563.07 +/- 173.40
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.90 +/- 0.28
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: sweep-v2
+        metrics:
+          - type: total_reward
+            value: 487.19 +/- 60.02
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.94 +/- 0.12
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: sweep-into-v2
+        metrics:
+          - type: total_reward
+            value: 798.80 +/- 15.62
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: window-close-v2
+        metrics:
+          - type: total_reward
+            value: 562.48 +/- 91.17
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.17
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Meta-World
+          type: window-open-v2
+        metrics:
+          - type: total_reward
+            value: 573.69 +/- 93.98
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.96 +/- 0.17
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandblockstack
+        metrics:
+          - type: total_reward
+            value: 347.40 +/- 50.60
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.17 +/- 0.23
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandbottlecap
+        metrics:
+          - type: total_reward
+            value: 338.25 +/- 81.25
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.81 +/- 0.25
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandcatchabreast
+        metrics:
+          - type: total_reward
+            value: 11.81 +/- 21.28
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.17 +/- 0.32
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandcatchover2underarm
+        metrics:
+          - type: total_reward
+            value: 31.60 +/- 7.20
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.92 +/- 0.24
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandcatchunderarm
+        metrics:
+          - type: total_reward
+            value: 18.21 +/- 9.46
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.72 +/- 0.39
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhanddoorcloseinward
+        metrics:
+          - type: total_reward
+            value: 3.97 +/- 0.15
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.36 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhanddoorcloseoutward
+        metrics:
+          - type: total_reward
+            value: 358.50 +/- 4.50
+            name: Total reward
+          - type: normalized_total_reward
+            value: -1.27 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhanddooropeninward
+        metrics:
+          - type: total_reward
+            value: 108.25 +/- 8.50
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.29 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhanddooropenoutward
+        metrics:
+          - type: total_reward
+            value: 83.65 +/- 12.10
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.13 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandgraspandplace
+        metrics:
+          - type: total_reward
+            value: 485.15 +/- 89.10
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.18
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandkettle
+        metrics:
+          - type: total_reward
+            value: -450.47 +/- 0.00
+            name: Total reward
+          - type: normalized_total_reward
+            value: -0.99 +/- 0.00
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandliftunderarm
+        metrics:
+          - type: total_reward
+            value: 377.92 +/- 13.24
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandover
+        metrics:
+          - type: total_reward
+            value: 33.01 +/- 0.96
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandpen
+        metrics:
+          - type: total_reward
+            value: 98.80 +/- 83.60
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.52 +/- 0.44
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandpushblock
+        metrics:
+          - type: total_reward
+            value: 445.60 +/- 2.20
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandreorientation
+        metrics:
+          - type: total_reward
+            value: 2798.00 +/- 2112.00
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.89 +/- 0.66
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandscissors
+        metrics:
+          - type: total_reward
+            value: 747.95 +/- 7.65
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.03 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandswingcup
+        metrics:
+          - type: total_reward
+            value: 3775.50 +/- 583.70
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.13
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandswitch
+        metrics:
+          - type: total_reward
+            value: 268.25 +/- 2.35
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Bi-DexHands
+          type: shadowhandtwocatchunderarm
+        metrics:
+          - type: total_reward
+            value: 2.17 +/- 0.67
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.03 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-0-v1
+        metrics:
+          - type: total_reward
+            value: -191.39 +/- 22.96
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.94 +/- 0.13
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-5-v1
+        metrics:
+          - type: total_reward
+            value: -194.01 +/- 3.66
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-10-v1
+        metrics:
+          - type: total_reward
+            value: -213.28 +/- 2.01
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.01 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-15-v1
+        metrics:
+          - type: total_reward
+            value: -227.82 +/- 4.29
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.01 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-20-v1
+        metrics:
+          - type: total_reward
+            value: -259.99 +/- 22.70
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.11
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-25-v1
+        metrics:
+          - type: total_reward
+            value: -282.28 +/- 20.70
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.11
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-30-v1
+        metrics:
+          - type: total_reward
+            value: -307.02 +/- 19.23
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.90 +/- 0.10
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-35-v1
+        metrics:
+          - type: total_reward
+            value: -314.36 +/- 5.62
+            name: Total reward
+          - type: normalized_total_reward
+            value: 1.00 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-40-v1
+        metrics:
+          - type: total_reward
+            value: -339.34 +/- 9.57
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.05
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-45-v1
+        metrics:
+          - type: total_reward
+            value: -366.63 +/- 7.47
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.04
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-50-v1
+        metrics:
+          - type: total_reward
+            value: -395.94 +/- 17.65
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.91 +/- 0.09
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-55-v1
+        metrics:
+          - type: total_reward
+            value: -403.73 +/- 2.03
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-60-v1
+        metrics:
+          - type: total_reward
+            value: -434.25 +/- 4.12
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.02
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-65-v1
+        metrics:
+          - type: total_reward
+            value: -480.31 +/- 8.63
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.86 +/- 0.04
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-70-v1
+        metrics:
+          - type: total_reward
+            value: -480.76 +/- 5.98
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.95 +/- 0.03
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-75-v1
+        metrics:
+          - type: total_reward
+            value: -476.83 +/- 2.44
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.99 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-80-v1
+        metrics:
+          - type: total_reward
+            value: -497.13 +/- 2.95
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.96 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-85-v1
+        metrics:
+          - type: total_reward
+            value: -513.83 +/- 3.06
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.98 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-90-v1
+        metrics:
+          - type: total_reward
+            value: -532.70 +/- 3.61
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-95-v1
+        metrics:
+          - type: total_reward
+            value: -557.42 +/- 3.81
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.01
+            name: Expert normalized total reward
+      - task:
+          type: in-context-reinforcement-learning
+          name: In-Context Reinforcement Learning
+        dataset:
+          name: Industrial-Benchmark
+          type: industrial-benchmark-100-v1
+        metrics:
+          - type: total_reward
+            value: -574.57 +/- 4.37
+            name: Total reward
+          - type: normalized_total_reward
+            value: 0.97 +/- 0.01
+            name: Expert normalized total reward
 ---
 # Model Card for Vintix