Metraipolis › archive › dc_56a7b747fa8a

Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Models — arXiv:2609.30048

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.30048
License (as archived)
CC0-1.0
Archived
2026-10-08T04:26:28.193Z
SHA-256 of the archived text
8cd8126247486c2c3c3911fb6525be38f4ab6765bad48539cf5c4ac649c95998
Stable URL of this page
https://metraipolis.com/documento/dc_56a7b747fa8a

How to cite

Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Models — arXiv:2609.30048 — arXiv (en), https://arxiv.org/abs/2609.30048 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_56a7b747fa8a (Metraipolis archive dc_56a7b747fa8a, sha256 8cd8126247486c2c3c3911fb6525be38f4ab6765bad48539cf5c4ac649c95998)
BibTeX
@misc{dc56a7b747fa8a,
  title = {Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Models — arXiv:2609.30048},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30048},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_56a7b747fa8a; sha256 8cd8126247486c2c3c3911fb6525be38f4ab6765bad48539cf5c4ac649c95998},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30048. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

If a language model can recognize code it wrote, it may favor that code as a judge, and instances of one model monitoring each other could collude. We test this zero-shot on current commercial models. Five LLMs generate solutions to MBPP, HumanEval, and DS-1000, seven more to MBPP, and models act as evaluators in four tasks: picking their own solution from a pair, judging whether a single solution is their own, identifying which of two solutions a named model wrote, and judging quality blind. In the single-solution task, balanced accuracy is 49-58% for all 15 model-benchmark combinations, while raw accuracy (38-67%) mostly reflects how readily a model claims authorship. In the pairwise task, accuracy across 14 evaluator-opponent combinations correlates at r=0.93 with how often the evaluator's solution is longer. Attribution to a named model succeeds on some pairs and is consistently inverted on others. A rule-based normalization that strips docstrings, comments, type hints, and local names preserves Pass@1 and leaves ten of twelve re-tested results at chance; the other two follow a length difference it leaves, although a trained classifier still separates most normalized pairs. Claude Haiku's self-preference also disappears. We recommend reporting balanced accuracy, heuristic baselines, and label consistency.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2609.30048 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T03:50:57.866Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).