SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.09067 · 2023

Holodeck: Language Guided Generation of 3D Embodied AI Environments

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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 0 have no contract verdict; 5 failures of our instrument; 0 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/2312.09067.json · badge: /api/ran/2312.09067.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
allenai/Holodeck canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
create_boundary_constraints Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
allenai/Holodeck/ai2holodeck/generation/milp_utils.py
code served (permissive licence) · get_code("e9608d34e71202fd")
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 cvxpy as cp
import numpy as np

# Construct arguments for create_boundary_constraints
c = cp.Variable(4)  # Variable for center position and rotation (cx, cy, ?, rotate_90)
object_dim = (2.0, 3.0)  # (x_size, y_size) - distinct sizes
bbox = (0.0, 0.0, 10.0, 10.0)  # (room_min_x, room_min_y, room_max_x, room_max_y)

ARGS = {
    "c": c,
    "object_dim": object_dim,
    "bbox": bbox
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cvxpy, numpy — 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
create_directional_constraints Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
allenai/Holodeck/ai2holodeck/generation/milp_utils.py
code served (permissive licence) · get_code("b9b5cdf6fd1c7295")
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 cvxpy as cp

ARGS = {
    "c1": [
        cp.Variable(),
        cp.Variable(),
        cp.Variable(boolean=True),
        cp.Variable(boolean=True),
    ],
    "c2": [
        cp.Variable(),
        cp.Variable(),
        cp.Variable(boolean=True),
        cp.Variable(boolean=True),
    ],
    "object_dim_1": (10.0, 20.0),
    "object_dim_2": (15.0, 25.0),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cvxpy — 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
create_nooverlap_constraints Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
allenai/Holodeck/ai2holodeck/generation/milp_utils.py
code served (permissive licence) · get_code("bf4f5436917acdd7")
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 cvxpy as cp
import torch
torch.manual_seed(0)

# Analyze the function signature:
# create_nooverlap_constraints(c1, c2, object_dim_1, object_dim_2)
#
# From the function body:
# - c1 and c2 are accessed with indices [0], [1], [3]
#   so they need to be indexable with at least 4 elements
# - c1[0], c1[1], c1[3] are used in cp.Variable operations
#   so they should be cp.Variable or numeric types
# - object_dim_1 and object_dim_2 are unpacked as (x_size, y_size)
#   so they need two elements each

# Create c1 and c2 as lists/tuples of cp.Variable objects
# c1 = [cx1, cy1, cz1, rotate_90_1]
c1 = [cp.Variable(), cp.Variable(), cp.Variable(), cp.Variable(boolean=True)]

# c2 = [cx2, cy2, cz2, rotate_90_2]
c2 = [cp.Variable(), cp.Variable(), cp.Variable(), cp.Variable(boolean=True)]

# object_dim_1 and object_dim_2 are (x_size, y_size) tuples
object_dim_1 = (10.0, 15.0)
object_dim_2 = (8.0, 12.0)

ARGS = {
    "c1": c1,
    "c2": c2,
    "object_dim_1": object_dim_1,
    "object_dim_2": object_dim_2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cvxpy, 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
get_room2layer Ran
ran; our contract draft was wrong, not the code
allenai/Holodeck/ai2holodeck/generation/layers.py
code served (permissive licence) · get_code("90209993f770d6a1")
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 torch
import numpy as np

# Analyze the function:
# - room_pairs: list of tuples/pairs representing room adjacencies
# - open_room_pairs: list of tuples/pairs representing open connections to remove
# Both are iterated over, and elements are indexed with [0] and [1]

# Create sample room pairs (adjacent rooms)
room_pairs = [
    (0, 1),
    (1, 2),
    (2, 3),
    (0, 3),
    (1, 3),
]

# Create sample open room pairs (connections to remove)
open_room_pairs = [
    (1, 3),
]

ARGS = {
    "room_pairs": room_pairs,
    "open_room_pairs": open_room_pairs,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
map_asset2layer Ran
ran; our contract draft was wrong, not the code
allenai/Holodeck/ai2holodeck/generation/layers.py
code served (permissive licence) · get_code("5b5c34c1a090e0cf")
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 torch
import numpy as np

# Construct a valid scene dictionary for map_asset2layer
scene = {
    'room_pairs': [
        ('room_0', 'room_1'),
        ('room_1', 'room_2'),
    ],
    'open_room_pairs': [],
    'rooms': [
        {'id': 'room_0'},
        {'id': 'room_1'},
        {'id': 'room_2'},
    ],
    'walls': [
        {'roomId': 'room_0'},
        {'roomId': 'room_1'},
    ],
    'objects': [
        {'roomId': 'room_0'},
        {'roomId': 'room_2'},
    ],
    'windows': [
        {'roomId': 'room_1'},
    ],
    'proceduralParameters': {
        'lights': [
            {'roomId': 'room_0'},
            {'roomId': 'room_1'},
        ]
    }
}

ARGS = {
    'scene': scene
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 5 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

3D simulated environments play a critical role in Embodied AI, but their creation requires expertise and extensive manual effort, restricting their diversity and scope. To mitigate this limitation, we present Holodeck, a system that generates 3D environments to match a user-supplied prompt fully automatedly. Holodeck can generate diverse scenes, e.g., arcades, spas, and museums, adjust the designs for styles, and can capture the semantics of complex queries such as "apartment for a researcher with a cat" and "office of a professor who is a fan of Star Wars". Holodeck leverages a large language model (i.e., GPT-4) for common sense knowledge about what the scene might look like and uses a large collection of 3D assets from Objaverse to populate the scene with diverse objects. To address the challenge of positioning objects correctly, we prompt GPT-4 to generate spatial relational constraints between objects and then optimize the layout to satisfy those constraints. Our large-scale human evaluation shows that annotators prefer Holodeck over manually designed procedural baselines in residential scenes and that Holodeck can produce high-quality outputs for diverse scene types. We also demonstrate an exciting application of Holodeck in Embodied AI, training agents to navigate in novel scenes like music rooms and daycares without human-constructed data, which is a significant step forward in developing general-purpose embodied agents.

For agents

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

get_harvested_code_for_paper("2312.09067")
get_code_for_paper("2312.09067")
have("2312.09067")

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

curl https://syntology.ai/api/ran/2312.09067.json

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

[![Syntology run record](https://syntology.ai/api/ran/2312.09067.svg)](https://syntology.ai/paper/2312.09067)

Connect an agent — have() is free.