SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.01906 · 2022

ASE: Large-Scale Reusable Adversarial Skill Embeddings for Physically Simulated Characters

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2205.01906.json · badge: /api/ran/2205.01906.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
nvlabs/protomotions community (classifier label)
classifier basis not recorded on the link
6 ran · 1 no recorded run
FunctionStatusWhere it lives
detect_checkpoint_mode Ran
ran on a synthesized input
nvlabs/protomotions/protomotions/train_agent.py
code served (permissive licence) · get_code("acb8928cac3540e6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
from pathlib import Path
import tempfile
import logging

logging.basicConfig(level=logging.INFO)

# Create a minimal args object with checkpoint attribute
class Args:
    def __init__(self):
        self.checkpoint = None

# Create temporary directory for testing
temp_dir = tempfile.mkdtemp()
save_dir = Path(temp_dir)

# Create args object
args = Args()

ARGS = {
    'args': args,
    'save_dir': save_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_experiment_module Ran
ran on a synthesized input
nvlabs/protomotions/protomotions/train_agent.py
code served (permissive licence) · get_code("ebd29fee3b781904")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned module. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import sys
from pathlib import Path
import tempfile
import textwrap

# Create a temporary experiment file
temp_dir = tempfile.mkdtemp()
experiment_file = Path(temp_dir) / "experiment.py"

# Write a minimal valid experiment module
experiment_code = textwrap.dedent("""
    # Minimal experiment module
    def run():
        return "experiment executed"
    
    config = {"learning_rate": 0.001}
""")

experiment_file.write_text(experiment_code)

# Define ARGS mapping all parameters to valid values
ARGS = {
    "experiment_path": str(experiment_file)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type module — a module output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a module output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_site_config Ran
ran on a synthesized input
nvlabs/protomotions/protomotions/train_slurm.py
code served (permissive licence) · get_code("ecb5faace2f48176")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
from pathlib import Path

# Create temporary directories and files for testing
tmpdir = tempfile.mkdtemp()
tmpdir_path = Path(tmpdir)

# Create a simple YAML site config file
site_config_content = """
login_node: 'test_login_node'
base_dir: '/test/experiments'
account: 'test_account'
partition: 'gpu'
container_mounts: '/scratch:/scratch:rw'
array_size: 5
"""

site_config_file = tmpdir_path / "slurm_site.yaml"
site_config_file.write_text(site_config_content)

# Create an alternative explicit path
alt_config_content = """
login_node: 'alt_login_node'
base_dir: '/alt/experiments'
account: 'alt_account'
"""

alt_config_file = tmpdir_path / "alt_site.yaml"
alt_config_file.write_text(alt_config_content)

# Create a config with clusters
cluster_config_content = """
account: 'shared_account'
default_cluster: 'gpu_cluster'
clusters:
  gpu_cluster:
    login_node: 'gpu.cluster.node'
    base_dir: '/gpu/experiments'
    partition: 'gpu'
  cpu_cluster:
    login_node: 'cpu.cluster.node'
    base_dir: '/cpu/experiments'
    partition: 'cpu'
"""

cluster_config_file = tmpdir_path / "cluster_site.yaml"
cluster_config_file.write_text(cluster_config_content)

# Set up environment variable for one test case
os.environ['PROTOMOTIONS_SLURM_SITE'] = str(alt_config_file)

ARGS = {
    'explicit_path': None,
    'cluster': None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
positive_int Ran
ran on a synthesized input
nvlabs/protomotions/protomotions/train_agent.py
code served (permissive licence) · get_code("7d0d7c9b4051de52")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import argparse

def positive_int(value):
    """Parse a strictly positive integer CLI value."""
    parsed = int(value)
    if parsed <= 0:
        raise argparse.ArgumentTypeError('must be greater than zero')
    return parsed

ARGS = {
    'value': '42'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
subprocess_run Ran
ran on a synthesized input
nvlabs/protomotions/protomotions/train_slurm.py
code served (permissive licence) · get_code("4fc05182f8ed8d64")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned CompletedProcess. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import subprocess

# The function subprocess_run takes:
# - cmd: a command to run (list or string)
# - ignore_err: boolean flag (default False)
# - **kwargs: additional keyword arguments to pass to subprocess.run()

# Looking at the function body:
# - cmd is passed directly to subprocess.run()
# - ignore_err is used in a boolean check
# - **kwargs are passed to subprocess.run()

# subprocess.run expects cmd to be a list of strings or a string
# Common kwargs include: capture_output, text, shell, etc.

# Create a simple, safe command that will succeed
# Using 'echo' with a simple message
cmd_value = ['echo', 'hello']

# Build the arguments dict
ARGS = {
    'cmd': cmd_value,
    'ignore_err': False,
    'capture_output': True,
    'text': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CompletedProcess — a CompletedProcess output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CompletedProcess output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
unconfigured_values Ran
ran on a synthesized input
nvlabs/protomotions/protomotions/train_slurm.py
code served (permissive licence) · get_code("9b497b94b6cd79cf")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

# Construct a site dict with various configuration values
# Some contain placeholder markers, some don't

site = {
    # Normal settings without placeholders
    "database": {
        "host": "localhost",
        "port": 5432
    },
    "debug": False,
    
    # Settings with placeholder markers that should be detected
    "api_key": "YOUR_API_KEY_HERE",
    "credentials": {
        "username": "your_account",
        "password": "secret123"
    },
    "backup_path": "/path/to/backup",
    "output_dir": {
        "location": "/path/to/output",
        "enabled": True
    },
    
    # Private settings (should be ignored)
    "_internal": {
        "YOUR_PLACEHOLDER": "should_be_ignored"
    },
    
    # Normal nested settings
    "server": {
        "name": "prod-server",
        "timeout": 30
    }
}

ARGS = {
    "site": site
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
resolve_asset_root Not yet run
unverified (harvested, no recorded run)
nvlabs/protomotions/protomotions/assets.py
code served (permissive licence) · get_code("014e3665ca489742")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The incredible feats of athleticism demonstrated by humans are made possible in part by a vast repertoire of general-purpose motor skills, acquired through years of practice and experience. These skills not only enable humans to perform complex tasks, but also provide powerful priors for guiding their behaviors when learning new tasks. This is in stark contrast to what is common practice in physics-based character animation, where control policies are most typically trained from scratch for each task. In this work, we present a large-scale data-driven framework for learning versatile and reusable skill embeddings for physically simulated characters. Our approach combines techniques from adversarial imitation learning and unsupervised reinforcement learning to develop skill embeddings that produce life-like behaviors, while also providing an easy to control representation for use on new downstream tasks. Our models can be trained using large datasets of unstructured motion clips, without requiring any task-specific annotation or segmentation of the motion data. By leveraging a massively parallel GPU-based simulator, we are able to train skill embeddings using over a decade of simulated experiences, enabling our model to learn a rich and versatile repertoire of skills. We show that a single pre-trained model can be effectively applied to perform a diverse set of new tasks. Our system also allows users to specify tasks through simple reward functions, and the skill embedding then enables the character to automatically synthesize complex and naturalistic strategies in order to achieve the task objectives.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2205.01906")
get_code_for_paper("2205.01906")
have("2205.01906")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2205.01906.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2205.01906.svg)](https://syntology.ai/paper/2205.01906)

Connect an agent — have() is free.