SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.04694 · ICML · 2025

EpiCoder: Encompassing Diversity and Complexity in Code Generation

Jie Wu, Qi Chen, Xin Zhang, Xiao Liu, Yujiu Yang, Yangyu Huang, Wenxiang Hu, Yaoxiang Wang, Haoling Li, Zhongxin Guo, Ying Xin, Jinsong Su, and 1 more

arXiv · PDF · Open in the Atlas

Code that ran

14 of 15 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 14 have no contract verdict; 0 failures of our instrument; 1 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/2501.04694.json · badge: /api/ran/2501.04694.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
DeepLearnXMU/EpiCoder canonical (classifier label)
classifier basis not recorded on the link
13 ran · 1 no recorded run
microsoft/EpiCoder canonical (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
custom_compare Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/extract/extract_frequency_count.py
pointer only (licence: MIT) · get_code("874482fd57d4cebe")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_content Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/debug/run_test_iter0.py
pointer only (licence: MIT) · get_code("e3c9ed0f508ba2ea")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_content Ran
ran on a synthesized input
microsoft/EpiCoder/gen/gen_code.py
code served (permissive licence) · get_code("d0f9fa36c5d6a449")
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 tuple. 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 re
import logging

# Set up logging to capture any errors
logging.basicConfig(level=logging.ERROR)

# Create a test solution string with JSON and code blocks
solution = """
<json>
{
  "packages": ["numpy", "pandas", "torch"],
  "file_names": ["main.py", "utils.py"]
}
</json>

<file>main.py</file>

import numpy as np

def hello():
    print("Hello from main")


<file>utils.py</file>

def utility_function(x):
    return x * 2

"""

ARGS = {
    "solution": solution
}
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 tuple — a tuple output has no array shape of its own; none is 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
extract_fields_from_formatted_text Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/gen/gen_question.py
pointer only (licence: MIT) · get_code("fd8dc3b17db50e82")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

form_feature_explanation Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/evol/feature_evol.py
pointer only (licence: MIT) · get_code("7237fedbb391f513")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_features Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/extract/construct_tree.py
pointer only (licence: MIT) · get_code("e6d8c07fb59f049e")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_full_content Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/extract/extract_features.py
pointer only (licence: MIT) · get_code("f46485f691c95440")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_large_files Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/utils/clear_code_debug_files.py
pointer only (licence: MIT) · get_code("6150c6d6a7ed7b25")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_question Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/gen/gen_code.py
pointer only (licence: MIT) · get_code("91751933774cda1d")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_text Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/evol/feature_evol.py
pointer only (licence: MIT) · get_code("6e1c2fcd1329e0bc")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

prune_to_top_n Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/extract/extract_frequency_count.py
pointer only (licence: MIT) · get_code("48e55621bbfcb75f")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

remove_keys Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/evol/feature_evol.py
pointer only (licence: MIT) · get_code("c1811a3b989ac815")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

shorten_repeated_substrings Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/call_api_local/call_api.py
pointer only (licence: MIT) · get_code("42f192ac4d7c7b09")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

sort_feature_frequency Ran
ran on a synthesized input
DeepLearnXMU/EpiCoder/extract/extract_frequency_count.py
pointer only (licence: MIT) · get_code("993fdd25aaa46031")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_data_from_file Not yet run
unverified (harvested, no recorded run)
DeepLearnXMU/EpiCoder/extract/extract_features.py
pointer only (licence: MIT) · get_code("3d9347bfa4f46252")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, DeepLearnXMU/EpiCoder, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 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 15 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

Existing methods for code generation use code snippets as seed data, restricting the complexity and diversity of the synthesized data. In this paper, we introduce a novel feature tree-based synthesis framework, which revolves around hierarchical code features derived from high-level abstractions of code. The feature tree is constructed from raw data and refined iteratively to increase the quantity and diversity of the extracted features, which captures and recognizes more complex patterns and relationships within the code. By adjusting the depth and breadth of the sampled subtrees, our framework provides precise control over the complexity of the generated code, enabling functionalities that range from function-level operations to multi-file scenarios. We fine-tuned widely-used base models to obtain EpiCoder series, achieving state-of-the-art performance on multiple benchmarks at both the function and file levels. In particular, empirical evidence indicates that our approach shows significant potential in the synthesizing of repository-level code data. Our code and data are publicly available. 1

For agents

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

get_harvested_code_for_paper("2501.04694")
get_code_for_paper("2501.04694")
have("2501.04694")

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

curl https://syntology.ai/api/ran/2501.04694.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.04694.svg)](https://syntology.ai/paper/2501.04694)

Connect an agent — have() is free.