Pith Number

pith:2QQIH7EN

pith:2026:2QQIH7EN5QKHJOWB3YZCLNP33W

not attested not anchored not stored refs resolved

From I/O to Code with Discovery Agent

Binhua Li, Ge Li, Haoran Zhang, Jiaru Qian, Peixu Wang, Xiaokang Yang, Xue Jiang, Yihong Dong, Yongbin Li, Zhi Jin

DIO-Agent synthesizes code from input-output examples by framing the task as evolutionary search with an LLM mutation operator guided by a simplicity-first bias.

arxiv:2605.15334 v1 · 2026-05-14 · cs.LG · cs.AI · cs.CL · cs.SE

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{2QQIH7EN5QKHJOWB3YZCLNP33W}

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

DIO-Agent consistently outperforms both traditional program-by-example method and SOTA evolution-agent baselines across all difficulty levels and various LLMs, while substantially surpassing test-time scaling strategies with equivalent sampling budgets.

C2weakest assumption

The Transformation Priority Premise successfully biases the LLM mutation operator toward the simplest hypothesis consistent with current evidence without missing valid complex solutions or introducing bias that harms search on harder instances.

C3one line summary

DIO-Agent frames IO2Code as LLM-driven evolutionary search over programs with a Transformation Priority Premise to favor simple hypotheses, outperforming baselines on a new IO2CodeBench.

References

28 extracted · 28 resolved · 5 Pith anchors

[1] Codeevolve: An open source evolutionary coding agent for algorithm discovery and optimization

[2] Language models are few-shot learners 1901

[3] Magellan: Autonomous discovery of novel compiler optimization heuristics with alphaevolve.arXiv preprint arXiv:2601.21096,

[4] Evaluating Large Language Models Trained on Code 2026 · arXiv:2107.03374

[5] Yihong Dong, Xue Jiang, Zhi Jin, and Ge Li · doi:10.1145/3425898.3426952

Formal links

1 machine-checked theorem link

Receipt and verification

First computed	2026-05-20T00:00:53.107753Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

d42083fc8dec1474bac1de3225b5fbddaf1f886deaf05f434d1756ff7c7e3143

Aliases

arxiv: 2605.15334 · arxiv_version: 2605.15334v1 · doi: 10.48550/arxiv.2605.15334 · pith_short_12: 2QQIH7EN5QKH · pith_short_16: 2QQIH7EN5QKHJOWB · pith_short_8: 2QQIH7EN

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/2QQIH7EN5QKHJOWB3YZCLNP33W \
  | 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: d42083fc8dec1474bac1de3225b5fbddaf1f886deaf05f434d1756ff7c7e3143

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "332d12c3605a28adf8475e45681b91e4d1050055b75cb0b1f647a466b28bbaaf",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CL",
      "cs.SE"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-14T18:57:32Z",
    "title_canon_sha256": "99a8409a4356379b635b0cdf36409395ff93f1a97c923de94f4cd5b584e18d6d"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15334",
    "kind": "arxiv",
    "version": 1
  }
}