Pith Number

pith:UCH5PWOF

pith:2023:UCH5PWOF7QIHGFDOPYMVRW2FTO

not attested not anchored not stored refs resolved

AgentCoder: Multi-Agent-based Code Generation with Iterative Testing and Optimisation

Dong Huang, Heming Cui, Jie M.Zhang, Michael Luck, Qingwen Bu, Yuhao Qing

A multi-agent system divides code generation among three agents to reach higher accuracy with lower token cost than single models.

arxiv:2312.13010 v3 · 2023-12-20 · cs.CL

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{UCH5PWOF7QIHGFDOPYMVRW2FTO}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

AgentCoder (GPT-4) achieves 96.3% and 91.8% pass@1 in HumanEval and MBPP datasets with an overall token overhead of 56.9K and 66.3K, while state-of-the-art obtains only 90.2% and 78.9% pass@1 with an overall token overhead of 138.2K and 206.5K.

C2weakest assumption

That iterative feedback from the test executor agent reliably improves code quality without introducing new errors or causing the programmer agent to overfit to the generated tests.

C3one line summary

A three-agent loop of code generation, test creation, and execution feedback lifts pass@1 to 96.3% on HumanEval and 91.8% on MBPP for GPT-4 while using roughly half the tokens of prior state-of-the-art.

References

54 extracted · 54 resolved · 17 Pith anchors

[1] Unified pre-training for program understanding and generation.arXiv preprint 2021

[2] Program Synthesis with Large Language Models 2021 · arXiv:2108.07732

[4] Language Models are Few-Shot Learners 2005 · arXiv:2005.14165

[5] arXiv preprint arXiv:2305.17126 , year= 2023

[7] Evaluating Large Language Models Trained on Code 2021 · arXiv:2107.03374

Formal links

3 machine-checked theorem links

Cited by

40 papers in Pith

Self-Refining Topology Optimization via an LLM-Based Multi-Agent Framework

Assessing, Exploiting, and Mitigating Syntactic Robustness Failures in LLM-Based Code Generation

MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering

Specification-Driven Code Translation Powered by Large Language Models: How Far Are We?

Adversarial Agent Collaboration for Correctness Improvements of C to Safe Rust Translation

Receipt and verification

First computed	2026-05-17T23:38:53.608051Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

a08fd7d9c5fc1073146e7e1958db459bb4659d19bbbd7e45013838433479baa2

Aliases

arxiv: 2312.13010 · arxiv_version: 2312.13010v3 · doi: 10.48550/arxiv.2312.13010 · pith_short_12: UCH5PWOF7QIH · pith_short_16: UCH5PWOF7QIHGFDO · pith_short_8: UCH5PWOF

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/UCH5PWOF7QIHGFDOPYMVRW2FTO \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: a08fd7d9c5fc1073146e7e1958db459bb4659d19bbbd7e45013838433479baa2

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f82e5695b3ec0d0a1c717f6d512cde914ec019994f33804d782a92a621d4e1c0",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2023-12-20T13:22:41Z",
    "title_canon_sha256": "5a6bba019c25f78396c3860172ae82837bc1075766dc395859feffa808ad737f"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2312.13010",
    "kind": "arxiv",
    "version": 3
  }
}