SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.10465 · 2022

SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualization

arXiv · PDF · Open in the Atlas

Code that ran

6 of 6 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, 1 violated it and 3 have no contract verdict; 2 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/2212.10465.json · badge: /api/ran/2212.10465.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
skywalker023/sodaverse canonical (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
FunctionStatusWhere it lives
capture_speakers Ran
ran on a synthesized input
skywalker023/sodaverse/utils/dialogue_utils.py
code served (permissive licence) · get_code("a21ca3fbcf9d2760")
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 re

def capture_speakers(dialogue):
    """
    Capture any string between the newline and the colon (:)
    """
    re_pattern = '(?<=\\n)(.*)(?=:)'
    speakers = re.findall(re_pattern, '\n' + dialogue)
    speakers = [s.strip() for s in speakers]
    return speakers

# The function takes a dialogue string and extracts speaker names
# that appear before colons in the format "Speaker:"
dialogue_example = """Alice: Hello, how are you?
Bob: I'm doing well, thanks for asking.
Alice: That's great to hear!
Charlie: Count me in too."""

ARGS = {
    "dialogue": dialogue_example
}
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 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
capture_utterances Ran
ran on a synthesized input
skywalker023/sodaverse/utils/dialogue_utils.py
code served (permissive licence) · get_code("5c7c0bb44a407ad3")
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 re

def capture_utterances(dialogue):
    """
    Capture any string between colon (:) and the newline
    """
    dialogue = dialogue.replace(':\n', ': ')
    re_pattern = '(?<=:)(.*)'
    utterances = re.findall(re_pattern, '\n' + dialogue)
    utterances = [u.strip() for u in utterances]
    return utterances


# Construct a valid dialogue string with speaker labels and utterances
dialogue = """Alice: Hello, how are you?
Bob: I'm doing well, thanks for asking.
Alice: That's great to hear!
Bob: What about you?"""

ARGS = {
    'dialogue': dialogue
}
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 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
check_built Ran
ran, violated the contract we drafted
skywalker023/sodaverse/tasks/dataset_loaders.py
code served (permissive licence) · get_code("1c14ee45c4283d31")
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 bool, ndim 0, dtype family float, 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 os
import tempfile
import shutil

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

# Create a .built file with version information
built_file_path = os.path.join(temp_dir, '.built')
with open(built_file_path, 'w') as f:
    f.write("some_build_info\n")
    f.write("v1.0.0\n")

# Define arguments for the check_built function
ARGS = {
    "path": temp_dir,
    "version_string": "v1.0.0"
}
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 bool, ndim 0, dtype family float, 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
deduplicate_sequent_chars Ran
ran on a synthesized input
skywalker023/sodaverse/utils/dialogue_utils.py
code served (permissive licence) · get_code("e007eeea174ded5d")
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 str. 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 re

def deduplicate_sequent_chars(string, chars):
    for c in chars:
        pattern = c + '{2,}'
        string = re.sub(pattern, c, string)
    return string

ARGS = {
    'string': 'aabbccddee',
    'chars': ['a', 'b', 'c', 'd', 'e']
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
download_and_check_hash Ran
ran; our contract draft was wrong, not the code
skywalker023/sodaverse/tasks/dataset_loaders.py
code served (permissive licence) · get_code("4ecf75a20f9f0944")
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 str. 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
import hashlib

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

# Create a small test file and compute its hash
test_data = b"This is a test file for hashing"
test_hash = hashlib.sha256(test_data).hexdigest()

# Create a mock response object that simulates requests.get behavior
class MockResponse:
    def __init__(self, data):
        self.data = data
        self.headers = {'content-length': str(len(data))}
        self.position = 0
    
    def iter_content(self, chunk_size=1024):
        for i in range(0, len(self.data), chunk_size):
            yield self.data[i:i+chunk_size]

# Patch requests.get to return our mock response
import requests
original_get = requests.get

def mock_get(url, stream=True):
    return MockResponse(test_data)

requests.get = mock_get

# Define the arguments
ARGS = {
    'url': 'https://example.com/testfile.bin',
    'filename': 'testfile.bin',
    'expected_hash': test_hash,
    'version': '1.0.0',
    'directory': temp_dir,
    'chunk_size': 1024 * 1024 * 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
requests, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
unzip_file Ran
ran; our contract draft was wrong, not the code
skywalker023/sodaverse/tasks/dataset_loaders.py
code served (permissive licence) · get_code("b3dded8ccf91020f")
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 str. 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 zipfile
import tarfile
import tempfile
import shutil

# Create a temporary directory for testing
test_dir = tempfile.mkdtemp()

# Create a test .zip file
zip_file_path = os.path.join(test_dir, 'test_archive.zip')
with zipfile.ZipFile(zip_file_path, 'w') as zf:
    zf.writestr('test_file.txt', 'test content')

# Create a test .tar.gz file
tar_file_path = os.path.join(test_dir, 'test_archive.tar.gz')
with tarfile.open(tar_file_path, 'w:gz') as tf:
    import io
    info = tarfile.TarInfo(name='test_file.txt')
    info.size = len(b'test content')
    tf.addfile(tarinfo=info, fileobj=io.BytesIO(b'test content'))

# Extract directory
extract_dir = os.path.join(test_dir, 'extract_dest')
os.makedirs(extract_dir, exist_ok=True)

ARGS = {
    'file_path': zip_file_path,
    'directory': extract_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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 3 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.

FunctionAlso present inEarliest-datedOthers named
check_built 1 other paper
1 later
2406.11911 (2024-06)
A Notion of Complexity for Theory of Mind via Discrete World Models
download_and_check_hash 1 other paper
1 later
2406.11911 (2024-06)
A Notion of Complexity for Theory of Mind via Discrete World Models
unzip_file 1 other paper
1 later
2406.11911 (2024-06)
A Notion of Complexity for Theory of Mind via Discrete World Models

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

Data scarcity has been a long standing issue in the field of open-domain social dialogue. To quench this thirst, we present SODA: the first publicly available, million-scale high-quality social dialogue dataset. By contextualizing social commonsense knowledge from a knowledge graph, we are able to distill an exceptionally broad spectrum of social interactions from a large language model. Human evaluation shows that conversations in SODA are more consistent, specific, and (surprisingly) natural than those in prior human-authored datasets. Using SODA, we train COSMO: a generalizable conversation model that is significantly more natural and consistent on unseen datasets than best-performing conversation models (e.g., GODEL, BlenderBot-1, Koala, Vicuna). Experiments reveal COSMO is sometimes even preferred to the original human-written gold responses. Additionally, our results shed light on the distinction between knowledge-enriched conversations and natural social chitchats. We plan to make our data, model, and code public.

For agents

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

get_harvested_code_for_paper("2212.10465")
get_code_for_paper("2212.10465")
have("2212.10465")

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

curl https://syntology.ai/api/ran/2212.10465.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.10465.svg)](https://syntology.ai/paper/2212.10465)

Connect an agent — have() is free.