SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.01780 · NeurIPS · 2022

CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning

Hung Le, Steven Hoi, Silvio Savarese, Yue Wang, Deepak Gotmare

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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 0 have no contract verdict; 1 failure 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/2207.01780.json · badge: /api/ran/2207.01780.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
salesforce/CodeRL canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
generate_prompt Ran
ran; our contract draft was wrong, not the code
salesforce/CodeRL/generate.py
code served (permissive licence) · get_code("6a5218b692e1a379")
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) 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 json
import os
import tempfile

torch.manual_seed(0)

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

# Create a prompt file
prompt_path = os.path.join(temp_dir, "prompt.txt")
with open(prompt_path, 'w') as f:
    f.write("Write a function that solves the problem.")

# Create a test case file with fn_name
test_case_path = os.path.join(temp_dir, "test_cases.json")
with open(test_case_path, 'w') as f:
    json.dump({"fn_name": "solve", "cases": []}, f)

# Create a solutions file
solutions_path = os.path.join(temp_dir, "solutions.txt")
with open(solutions_path, 'w') as f:
    f.write("def solve():\n    pass")

# Create a starter file
starter_path = os.path.join(temp_dir, "starter.py")
with open(starter_path, 'w') as f:
    f.write("def solve():\n    # Your code here\n    pass")

# Create a minimal tokenizer stand-in
class MinimalTokenizer:
    def __init__(self):
        self.vocab_size = 1000
    
    def encode(self, text):
        return [1, 2, 3]
    
    def decode(self, tokens):
        return "decoded text"

tokenizer = MinimalTokenizer()

# Create a minimal args object
class Args:
    def __init__(self):
        self.model_name = "test_model"
        self.max_length = 512

args = Args()

ARGS = {
    "args": args,
    "test_case_path": test_case_path,
    "prompt_path": prompt_path,
    "solutions_path": solutions_path,
    "tokenizer": tokenizer,
    "starter_path": starter_path,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 0 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
generate_prompt 1 other paper
1 later
2307.04349 (2023-07)
RLTF: Reinforcement Learning from Unit Test Feedback

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

Program synthesis or code generation aims to generate a program that satisfies a problem specification. Recent approaches using large-scale pretrained language models (LMs) have shown promising results, yet they have some critical limitations. In particular, they often follow a standard supervised fine-tuning procedure to train a code generation model only from the pairs of natural-language problem descriptions and ground-truth programs. Such paradigm largely ignores some important but potentially useful signals in the problem specification such as unit tests, which thus often results in poor performance when solving complex unseen coding tasks. To address the limitations, we propose "CodeRL", a new framework for program synthesis tasks through pretrained LMs and deep reinforcement learning (RL). Specifically, during training, we treat the code-generating LM as an actor network, and introduce a critic network that is trained to predict the functional correctness of generated programs and provide dense feedback signals to the actor. During inference, we introduce a new generation procedure with a critical sampling strategy that allows a model to automatically regenerate programs based on feedback from example unit tests and critic scores. For the model backbones, we extended the encoder-decoder architecture of CodeT5 with enhanced learning objectives, larger model sizes, and better pretraining data. Our method not only achieves new SOTA results on the challenging APPS benchmark, but also shows strong zero-shot transfer capability with new SOTA results on the simpler MBPP benchmark.

For agents

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

get_harvested_code_for_paper("2207.01780")
get_code_for_paper("2207.01780")
have("2207.01780")

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

curl https://syntology.ai/api/ran/2207.01780.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.01780.svg)](https://syntology.ai/paper/2207.01780)

Connect an agent — have() is free.