SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.02183 · 2024

Self-Organized Agents: A LLM Multi-Agent Framework toward Ultra Large-Scale Code Generation and Optimization

arXiv · PDF · Open in the Atlas

Code that ran

8 of 11 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 3 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/2404.02183.json · badge: /api/ran/2404.02183.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
tsukushiai/self-organized-agent canonical (classifier label)
classifier basis: census
8 ran · 3 no recorded run
FunctionStatusWhere it lives
count_test_cases Ran
ran, honoured the contract we drafted
tsukushiai/self-organized-agent/programming_runs/validate_py_results.py
code served (permissive licence) · get_code("8e4c36e8b56b0026")
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 torch

def count_test_cases(test_str: str) -> int:
    return test_str.count('assert')

# Construct valid arguments
test_str_value = """
def test_addition():
    assert 1 + 1 == 2
    assert 2 + 2 == 4

def test_subtraction():
    assert 5 - 3 == 2
    assert 10 - 4 == 6
    assert 7 - 7 == 0
"""

ARGS = {
    "test_str": test_str_value
}
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 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
function_with_timeout Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
tsukushiai/self-organized-agent/programming_runs/executors/executor_utils.py
code served (permissive licence) · get_code("02656b1924cbfdfb")
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 torch
from threading import Thread

class PropagatingThread(Thread):

    def run(self):
        self.exc = None
        try:
            if hasattr(self, '_Thread__target'):
                self.ret = self._Thread__target(*self._Thread__args, **self._Thread__kwargs)
            else:
                self.ret = self._target(*self._args, **self._kwargs)
        except BaseException as e:
            self.exc = e

    def join(self, timeout=None):
        super(PropagatingThread, self).join(timeout)
        if self.exc:
            raise self.exc
        return self.ret

def function_with_timeout(func, args, timeout):
    result_container = []

    def wrapper():
        result_container.append(func(*args))
    thread = PropagatingThread(target=wrapper)
    thread.start()
    thread.join(timeout)
    if thread.is_alive():
        raise TimeoutError()
    else:
        return result_container[0]


def simple_function(x, y):
    """A simple test function that returns the sum of arguments"""
    return x + y


ARGS = {
    'func': simple_function,
    'args': (5, 3),
    'timeout': 2.0
}
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 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
green_text Ran
ran; our contract draft was wrong, not the code
tsukushiai/self-organized-agent/programming_runs/validate_py_results.py
code served (permissive licence) · get_code("405659bf95a96f60")
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 torch

def green_text(text: str) -> str:
    return f'\x1b[92m{text}\x1b[0m'

ARGS = {
    'text': 'Hello, World!'
}
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 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
make_printv Ran
ran; our contract draft was wrong, not the code
tsukushiai/self-organized-agent/programming_runs/utils.py
code served (permissive licence) · get_code("b2c53dc3cc9993aa")
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 function. 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

torch.manual_seed(0)

ARGS = {
    'verbose': True
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
message_to_str Ran
ran; our contract draft was wrong, not the code
tsukushiai/self-organized-agent/programming_runs/generators/model.py
code served (permissive licence) · get_code("a067a230397d97aa")
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 torch
import dataclasses
from typing import Literal

MessageRole = Literal['system', 'user', 'assistant']

@dataclasses.dataclass()
class Message:
    role: MessageRole
    content: str

def message_to_str(message: Message) -> str:
    return f'{message.role}: {message.content}'

# Create a valid Message instance
test_message = Message(
    role='user',
    content='Hello, how are you?'
)

ARGS = {
    'message': test_message
}
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 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
messages_to_str Ran
ran; our contract draft was wrong, not the code
tsukushiai/self-organized-agent/programming_runs/generators/model.py
code served (permissive licence) · get_code("4e0f76b9ec8bd13a")
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 dataclasses
from typing import List, Literal

MessageRole = Literal['system', 'user', 'assistant']

@dataclasses.dataclass()
class Message:
    role: MessageRole
    content: str

def message_to_str(message: Message) -> str:
    return f'{message.role}: {message.content}'

def messages_to_str(messages: List[Message]) -> str:
    return '\n'.join([message_to_str(message) for message in messages])

# Construct valid arguments
messages = [
    Message(role='system', content='You are a helpful assistant.'),
    Message(role='user', content='What is 2+2?'),
    Message(role='assistant', content='The answer is 4.')
]

ARGS = {
    'messages': messages
}
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
read_jsonl_gz Ran
ran; our contract draft was wrong, not the code
tsukushiai/self-organized-agent/programming_runs/utils.py
code served (permissive licence) · get_code("ff9702cd987f980c")
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 gzip
import json
import tempfile
import os

# Create a temporary jsonl.gz file with sample data
temp_dir = tempfile.mkdtemp()
temp_file_path = os.path.join(temp_dir, "test_data.jsonl.gz")

sample_data = [
    {"id": 1, "text": "hello world"},
    {"id": 2, "text": "foo bar"},
    {"id": 3, "text": "test data"}
]

with gzip.open(temp_file_path, 'wt') as f:
    for item in sample_data:
        f.write(json.dumps(item) + '\n')

ARGS = {
    "path": temp_file_path
}
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
red_text Ran
ran; our contract draft was wrong, not the code
tsukushiai/self-organized-agent/programming_runs/validate_py_results.py
code served (permissive licence) · get_code("588974da3369998d")
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 torch

torch.manual_seed(0)

ARGS = {
    "text": "Hello, World!"
}
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 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
generate_code Not yet run
unverified (harvested, no recorded run)
tsukushiai/self-organized-agent/programming_runs/soas/agent.py
code served (permissive licence) · get_code("c4c0bb9c81930ee1")
How it ran

no invocation recorded.

generate_skeleton Not yet run
unverified (harvested, no recorded run)
tsukushiai/self-organized-agent/programming_runs/soas/agent.py
code served (permissive licence) · get_code("fa0e7ed5f7388f5e")
How it ran

no invocation recorded.

get_subtasks Not yet run
unverified (harvested, no recorded run)
tsukushiai/self-organized-agent/programming_runs/soas/agent.py
code served (permissive licence) · get_code("3f8d840db9fde8b2")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 5 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 5 later) and 6 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
message_to_str 4 other papers
2 dated earlier, 2 later
2310.04406 (2023-10)
Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models
2402.16906 (2024-02)
Debug like a Human: A Large Language Model Debugger via Verifying Runtime Execution Step-by-step
2407.11770 (2024-07)
Robust Utility-Preserving Text Anonymization Based on Large Language Models
and 1 more not named here
messages_to_str 4 other papers
2 dated earlier, 2 later
2310.04406 (2023-10)
Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models
2402.16906 (2024-02)
Debug like a Human: A Large Language Model Debugger via Verifying Runtime Execution Step-by-step
2407.11770 (2024-07)
Robust Utility-Preserving Text Anonymization Based on Large Language Models
and 1 more not named here
function_with_timeout 1 other paper
1 later
2407.09007 (2024-07)
Benchmarking Language Model Creativity: A Case Study on Code Generation
make_printv 1 other paper
1 later
2407.11770 (2024-07)
Robust Utility-Preserving Text Anonymization Based on Large Language Models
read_jsonl_gz 1 other paper
1 later
2407.11770 (2024-07)
Robust Utility-Preserving Text Anonymization Based on Large Language 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

Recent advancements in automatic code generation using large language model (LLM) agent have brought us closer to the future of automated software development. However, existing single-agent approaches face limitations in generating and improving large-scale, complex codebases due to constraints in context length. To tackle this challenge, we propose Self-Organized multi-Agent framework (SoA), a novel multi-agent framework that enables the scalable and efficient generation and optimization of large-scale code. In SoA, self-organized agents operate independently to generate and modify code components while seamlessly collaborating to construct the overall codebase. A key feature of our framework is the automatic multiplication of agents based on problem complexity, allowing for dynamic scalability. This enables the overall code volume to be increased indefinitely according to the number of agents, while the amount of code managed by each agent remains constant. We evaluate SoA on the HumanEval benchmark and demonstrate that, compared to a single-agent system, each agent in SoA handles significantly less code, yet the overall generated code is substantially greater. Moreover, SoA surpasses the powerful single-agent baseline by 5% in terms of Pass@1 accuracy.

For agents

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

get_harvested_code_for_paper("2404.02183")
get_code_for_paper("2404.02183")
have("2404.02183")

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

curl https://syntology.ai/api/ran/2404.02183.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.02183.svg)](https://syntology.ai/paper/2404.02183)

Connect an agent — have() is free.