Navers lab

EmbodiedEvo

Self-Evolving Coding Agent For Physical Intelligence

Through code, act and evolve in the physical world.

Navers Lab · Einsia AI · 2026

EmbodiedEvo overview: a coding agent develops Skills and Lessons through Playground feedback; a frozen harness is reused by fresh Test agents in RoboDojo

Method

Explore the full method →
Two-stage EmbodiedEvo framework: Playground cycles through code execution, physical feedback, and Skills and Lessons revision; frozen knowledge is then reused by fresh Test agents, with local code adaptation, read-only experience, and no reset.

Results

Evaluation setup & full results →

RoboDojo leaderboard

Snapshot · Oct 9, 2026Official leaderboard
RoboDojo simulation: seven systems from the paper's Oct 9, 2026 snapshot. Each metric cell lists Score / success rate. Higher is better; evaluation protocols differ across systems.
Model / contributorTypeScore/SRScore/SRScore/SRScore/SRScore/SRScore/SR
EmbodiedEvoOurs
Navers Lab · Einsia AIAgent
 Agent
44.48/37.82%
34.12/29.67%
23.65/13.25%
37.20/23.50%
71.90/69.67%
55.52/53.00%
VPP2
RobotEraWAM
 WAM
39.26/32.26%
36.38/29.50%
43.03/35.17%
56.28/37.00%
56.01/55.56%
4.61/4.08%
PhysicalRSI
HKU MMLab & KAIAgent + VLA
 Agent + VLA
36.27/31.38%
21.30/15.62%
38.05/32.50%
46.37/37.33%
46.74/46.56%
28.88/24.92%
Awomo-0.5
AwomoWAM
 WAM
35.34/29.64%
30.31/23.84%
39.65/33.50%
40.36/26.75%
42.27/41.22%
24.09/22.92%
Simate-beta
SimateVLA
 VLA
33.95/27.96%
35.09/27.95%
34.35/26.92%
57.84/43.42%
33.33/33.00%
9.12/8.50%
InternW0-Δ
Shanghai AI LabWAM
 WAM
30.77/23.91%
30.09/22.78%
31.98/23.25%
46.29/29.33%
34.67/34.00%
10.84/10.17%
GPT-6-Astra
RoboDojo TeamAgent
 Agent
28.97/22.48%
33.36/30.50%
12.65/4.00%
21.45/8.25%
43.04/38.67%
34.36/31.00%
↑ Higher is better · Score / success rate (%)Scroll to compare all capabilities

Key Findings

Explore all findings & evidence →
8.50 → 30.17%

Experience improves physical execution.

Astra/xhigh success on matched scenes with task descriptions alone. Gains persist when both agents also receive the wiki.

Paired results →
+8.86 / +5.50 pp

Knowledge transfers across agents.

Sol / Luna success-rate gains from Astra’s frozen experience, even when their own exploration brings no completion gains.

Transfer results →
0 → 40%

Unfinished exploration can still help.

Conveyor pickup success on five new scenes, using experience from a Playground task the agent never completed.

What the agent learned →

See it in action

Browse all trajectories →