SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.08992 · ICLR · 2024

CodeChain: Towards Modular Code Generation Through Chain of Self-revisions with Representative Sub-modules

Hung Le, Shafiq Joty, Akash Gokul, Hailin Chen, Amrita Saha, Doyen Sahoo

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has 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/2310.08992.json · badge: /api/ran/2310.08992.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
SalesforceAIResearch/CodeChain canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
BaseEncoder Ran
ran on a synthesized input · constructed an object; no result was computed
SalesforceAIResearch/CodeChain/src/embedding/encoder.py
code served (permissive licence) · get_code("f34f883bb6fb5e60")
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 an instance of its own class, BaseEncoder: the run constructed the object and did not call it, so no result was computed. 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 torch.nn as nn
from unittest.mock import Mock, MagicMock

torch.manual_seed(0)

# Create a minimal mock tokenizer
class MockTokenizer:
    def __init__(self):
        self.pad_token = '<pad>'
        self.sep_token = '<sep>'
        self.cls_token = '<cls>'
        self.mask_token = '<mask>'
    
    def add_special_tokens(self, tokens_dict):
        pass
    
    def __call__(self, text, **kwargs):
        return {
            'input_ids': torch.tensor([[1, 2, 3]]),
            'attention_mask': torch.tensor([[1, 1, 1]])
        }

# Create a minimal mock encoder model
class MockEncoder(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(10, 8)
    
    def forward(self, input_ids, attention_mask=None, **kwargs):
        # Return a mock last_hidden_state
        batch_size = input_ids.shape[0] if isinstance(input_ids, torch.Tensor) else len(input_ids)
        return Mock(last_hidden_state=torch.randn(batch_size, 3, 8))
    
    def eval(self):
        return self
    
    def to(self, device):
        return self

# Mock the AutoTokenizer.from_pretrained
original_auto_tokenizer = None
try:
    from transformers import AutoTokenizer as OriginalAutoTokenizer
    original_auto_tokenizer = OriginalAutoTokenizer.from_pretrained
except:
    pass

def mock_tokenizer_from_pretrained(model_name, **kwargs):
    return MockTokenizer()

def mock_model_from_pretrained(model_name, **kwargs):
    return MockEncoder()

# Patch the imports
import transformers
transformers.AutoTokenizer.from_pretrained = mock_tokenizer_from_pretrained
transformers.AutoModel.from_pretrained = mock_model_from_pretrained

ARGS = {
    'device': 'cpu',
    'max_input_len': 256,
    'maximum_token_len': 512,
    'model_name': 'bert-base-uncased'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type BaseEncoder — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a BaseEncoder 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
prepare_tokenizer Ran
ran; our contract draft was wrong, not the code
SalesforceAIResearch/CodeChain/src/embedding/encoder.py
code served (permissive licence) · get_code("55cffac4dad16773")
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 Mock. 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
from unittest.mock import Mock, MagicMock
from transformers import AutoTokenizer

torch.manual_seed(0)

# Create a mock tokenizer that behaves like AutoTokenizer.from_pretrained
mock_tokenizer = Mock(spec=AutoTokenizer)
mock_tokenizer.add_special_tokens = Mock(return_value=0)

# Mock the AutoTokenizer.from_pretrained to return our mock tokenizer
AutoTokenizer.from_pretrained = Mock(return_value=mock_tokenizer)

ARGS = {
    'tokenizer_path': 'bert-base-uncased'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Mock — a Mock output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Mock 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
truncate_sentences Ran
ran; our contract draft was wrong, not the code
SalesforceAIResearch/CodeChain/src/embedding/encoder.py
code served (permissive licence) · get_code("7396327c9cd9f836")
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
import numpy as np
from typing import List, Dict, Union

torch.manual_seed(0)

# Construct arguments for truncate_sentences
sentence_list = [
    "This is the first sentence.",
    "Another sentence with more words.",
    "Short one.",
    "The quick brown fox jumps over the lazy dog."
]

maximum_length = 10

ARGS = {
    "sentence_list": sentence_list,
    "maximum_length": maximum_length
}
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 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 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.

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

Large Language Models (LLMs) have already become quite proficient at solving simpler programming tasks like those in HumanEval or MBPP benchmarks. However, solving more complex and competitive programming tasks is still quite challenging for these models -possibly due to their tendency to generate solutions as monolithic code blocks instead of decomposing them into logical sub-tasks and sub-modules. On the other hand, experienced programmers instinctively write modularized code with abstraction for solving complex tasks, often reusing previously developed modules. To address this gap, we propose CodeChain, a novel framework for inference that elicits modularized code generation through a chain of self-revisions, each being guided by some representative sub-modules generated in previous iterations. Concretely, CodeChain first instructs the LLM to generate modularized codes through chain-of-thought prompting. Then it applies a chain of self-revisions by iterating the two steps: 1) extracting and clustering the generated sub-modules and selecting the cluster representatives as the more generic and re-usable implementations, and 2) augmenting the original chain-of-thought prompt with these selected module-implementations and instructing the LLM to re-generate new modularized solutions. We find that by naturally encouraging the LLM to reuse the previously developed and verified sub-modules, CodeChain can significantly boost both modularity as well as correctness of the generated solutions, achieving relative pass@1 improvements of 35% on APPS and 76% on CodeContests. It is shown to be effective on both OpenAI LLMs as well as opensourced LLMs like WizardCoder. We also conduct comprehensive ablation studies with different methods of prompting, number of clusters, model sizes, program qualities, etc., to provide useful insights that underpin CodeChain's success 1 .

For agents

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

get_harvested_code_for_paper("2310.08992")
get_code_for_paper("2310.08992")
have("2310.08992")

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

curl https://syntology.ai/api/ran/2310.08992.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.08992.svg)](https://syntology.ai/paper/2310.08992)

Connect an agent — have() is free.