SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.08127 · ACL · 2023

CB2: Collaborative Natural Language Interaction Research Platform

Yoav Artzi, Jacob Sharf, Mustafa Gul

arXiv · PDF · Open in the Atlas

Code that ran

7 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 7 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/2303.08127.json · badge: /api/ran/2303.08127.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
lil-lab/cb2 canonical (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
FunctionStatusWhere it lives
degree_to_rotation Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/models/rotation.py
code served (permissive licence) · get_code("2ba24bf1c744f445")
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 Rotation. 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 numpy as np

# The function degree_to_rotation takes a single parameter: degree (int)
# It maps integer degree values to Rotation enum values
# Valid inputs are: 30, 90, 150, 210, 270, 330

ARGS = {
    'degree': 30
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Rotation — a Rotation output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Rotation 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
get_cartesian Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/models/hex_conv.py
code served (permissive licence) · get_code("1850aaa151882405")
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 ndarray, ndim 1, dtype family float, shape (2,) 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 numpy as np

# The function get_cartesian takes 'coords' as input
# It expects coords to be array-like (list, tuple, or numpy array)
# that can be converted to a numpy array and used with np.matmul
# The basis_matrix is 2x2, so coords should be a 2-element vector or 2xN array

# Create a simple 2D coordinate or batch of coordinates
coords = np.array([1.0, 2.0])

ARGS = {
    "coords": coords
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) 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
get_distance Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/models/hex_conv.py
code served (permissive licence) · get_code("d10445c0642a87eb")
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 float64, ndim 0, 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 numpy as np

# Define coordinates for hexagonal lattice
# The function expects coordinates as input
target_coords = [1, 2]  # hexagonal coordinates for target
center_coords = [3, 1]  # hexagonal coordinates for center

ARGS = {
    "target": target_coords,
    "center": center_coords
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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_arguments Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/utils.py
code served (permissive licence) · get_code("5409c0b48b1c18f3")
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 json
import os
import tempfile

# Create a temporary directory with a valid args.json file
temp_dir = tempfile.mkdtemp()
args_file = os.path.join(temp_dir, 'args.json')

# Write a sample args.json file
sample_args = {
    "learning_rate": 0.001,
    "batch_size": 32,
    "epochs": 10,
    "model_name": "test_model"
}

with open(args_file, 'w') as f:
    json.dump(sample_args, f)

ARGS = {
    "args_dir": temp_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 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
load_json Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/utils.py
code served (permissive licence) · get_code("b22517f93442b71c")
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 json
import tempfile
import os

# Create a temporary JSON file with valid data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.json")

test_data = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": {"nested": "object"}
}

with open(temp_file, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "filename": temp_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 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
load_pickle Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/utils.py
code served (permissive licence) · get_code("bb951664cfdfafb9")
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 pickle
import tempfile
import os

# Create a temporary pickle file with some test data
test_data = {
    'key1': 'value1',
    'key2': [1, 2, 3],
    'key3': {'nested': 'dict'}
}

# Create a temporary file and write pickle data to it
temp_fd, temp_path = tempfile.mkstemp(suffix='.pkl')
os.close(temp_fd)

try:
    with open(temp_path, 'wb') as f:
        pickle.dump(test_data, f)
    
    ARGS = {
        'filename': temp_path
    }
finally:
    # Clean up will happen after execution
    pass
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 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
rotation_from_v3 Ran
ran on a synthesized input
lil-lab/cb2/follower_bots/models/rotation.py
code served (permissive licence) · get_code("7edfc5575995533a")
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 Rotation. 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.

from enum import Enum
import numpy as np

class Rotation(str, Enum):
    """Rotations in the environment."""
    NORTHEAST: str = 'NORTHEAST'
    NORTHWEST: str = 'NORTHWEST'
    SOUTHEAST: str = 'SOUTHEAST'
    SOUTHWEST: str = 'SOUTHWEST'
    EAST: str = 'EAST'
    WEST: str = 'WEST'

    def shorthand(self) -> str:
        if self == Rotation.NORTHEAST:
            return 'NE'
        elif self == Rotation.EAST:
            return 'EAST'
        elif self == Rotation.SOUTHEAST:
            return 'SE'
        elif self == Rotation.SOUTHWEST:
            return 'SW'
        elif self == Rotation.WEST:
            return 'WEST'
        elif self == Rotation.NORTHWEST:
            return 'NW'
        else:
            raise ValueError('Could not convert: ' + str(self))

    def __str__(self):
        return self.value

    def __hash__(self):
        return self.value.__hash__()

    def __int__(self) -> int:
        if self == Rotation.NORTHEAST:
            return 30
        elif self == Rotation.EAST:
            return 90
        elif self == Rotation.SOUTHEAST:
            return 150
        elif self == Rotation.SOUTHWEST:
            return 210
        elif self == Rotation.WEST:
            return 270
        elif self == Rotation.NORTHWEST:
            return 330
        else:
            raise ValueError('Could not convert: ' + str(self))

    def to_radians(self) -> float:
        offset_deg: int = (int(self) - 90 + 360) % 360
        return np.deg2rad(offset_deg)

    def to_v3(self) -> str:
        return f'0,{int(self)},0'

def degree_to_rotation(degree: int) -> Rotation:
    """Maps from an integer degree rotation given by Unity to a Rotation type.

    Input:
        degree (int): An integer degree rotation from Unity.

    Returns:
        The corresponding Rotation.

    Raises:
        ValueError, if the degree integer is not in the range (30 + 60x), where x is an integer in [0, 5].
    """
    if degree == 30:
        return Rotation.NORTHEAST
    elif degree == 90:
        return Rotation.EAST
    elif degree == 150:
        return Rotation.SOUTHEAST
    elif degree == 210:
        return Rotation.SOUTHWEST
    elif degree == 270:
        return Rotation.WEST
    elif degree == 330:
        return Rotation.NORTHWEST
    else:
        raise ValueError('Degree not in 30+60x; ' + str(degree))

def rotation_from_v3(v3: str) -> Rotation:
    return degree_to_rotation(int(float(eval(v3)[1])))

ARGS = {
    'v3': '0,90,0'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Rotation — a Rotation output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Rotation 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 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

CB2 is a multi-agent platform to study collaborative natural language interaction in a grounded task-oriented scenario. It includes a 3D game environment, a backend server designed to serve trained models to human agents, and various tools and processes to enable scalable studies. We deploy CB2 at https://cb2.ai as a system demonstration with a learned instruction following model.

For agents

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

get_harvested_code_for_paper("2303.08127")
get_code_for_paper("2303.08127")
have("2303.08127")

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

curl https://syntology.ai/api/ran/2303.08127.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.08127.svg)](https://syntology.ai/paper/2303.08127)

Connect an agent — have() is free.