Metraipolis › archive › dc_ba67d72b6cbd

Weight-space programming for auditing clinical decision networks — arXiv:2611.05341

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.05341
License (as archived): CC0-1.0
Archived: 2026-11-04T21:40:29.692Z
SHA-256 of the archived text: ee62566f4f56e2f853faf3a69d54e92e5ab6fbdfe5074356cfbe0c96aab69181
Stable URL of this page: https://metraipolis.com/documento/dc_ba67d72b6cbd

How to cite

Weight-space programming for auditing clinical decision networks — arXiv:2611.05341 — arXiv (en), https://arxiv.org/abs/2611.05341 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ba67d72b6cbd (Metraipolis archive dc_ba67d72b6cbd, sha256 ee62566f4f56e2f853faf3a69d54e92e5ab6fbdfe5074356cfbe0c96aab69181)

BibTeX

@misc{dcba67d72b6cbd,
  title = {Weight-space programming for auditing clinical decision networks — arXiv:2611.05341},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.05341},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ba67d72b6cbd; sha256 ee62566f4f56e2f853faf3a69d54e92e5ab6fbdfe5074356cfbe0c96aab69181},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05341. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine learning models in radiotherapy are commissioned by testing them on cases with known answers, which may not reveal a shortcut that the test cases share with the training data. Frontier language models have recently been reported to read and write the raw weights of small neural networks, an ability called weight-space programming. The purpose of this study was to determine whether such a model, given only a clinical protocol and a network's weights, can audit and repair the network. Networks with 433 to 4,801 parameters were trained to screen head-and-neck planning requests against four rules. Flaws were planted through the training data: confounds that hid each rule, a 3 Gy shift in the cord threshold, and a hidden trigger. Claude Opus 5.5 audited each network blind, without tools, data or code execution, and every claim and repair was tested on about 420,000 requests. The language model named the broken rule in 16 of 18 flawed networks (95% CI 67% to 97%), including all 15 with a confound or a threshold shift. It repaired each of the 16 by changing a median of 3 parameters. Fine-tuning on protocol-labeled requests also repaired every network, but changed nearly every parameter. Each confound and shift it found was visible in single units, and a simple statistic chosen for each flaw type found the same flaws. It cleared 2 of 5 healthy networks without being forced and found 1 of 3 triggers. Testing against the protocol detected every flaw, and black-box testing found the confounds and the shift faster. The results indicate that, in small networks, weight-space programming can locate and repair flaws that silence an input or shift a threshold, but cannot yet certify that a network is correct.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Medical Physics (physics.med-ph) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.05341 — 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 2026-11-04T21:35:49.725Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).