Metraipolis › archive › dc_8bb0fddd6466

Beyond End-to-End Recovery: Towards a Better Understanding of LLM Steganography via Rolewise Decomposition — arXiv:2610.30516

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.30516
License (as archived): CC0-1.0
Archived: 2026-10-24T10:48:25.197Z
SHA-256 of the archived text: a33d926289acbc0f75a80b770bc36094e9cf0e62a3827ac017b9d7e55e28a757
Stable URL of this page: https://metraipolis.com/documento/dc_8bb0fddd6466

How to cite

Beyond End-to-End Recovery: Towards a Better Understanding of LLM Steganography via Rolewise Decomposition — arXiv:2610.30516 — arXiv (en), https://arxiv.org/abs/2610.30516 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8bb0fddd6466 (Metraipolis archive dc_8bb0fddd6466, sha256 a33d926289acbc0f75a80b770bc36094e9cf0e62a3827ac017b9d7e55e28a757)

BibTeX

@misc{dc8bb0fddd6466,
  title = {Beyond End-to-End Recovery: Towards a Better Understanding of LLM Steganography via Rolewise Decomposition — arXiv:2610.30516},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30516},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8bb0fddd6466; sha256 a33d926289acbc0f75a80b770bc36094e9cf0e62a3827ac017b9d7e55e28a757},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30516. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In large language model (LLM) steganography, an encoder hides a private message in an otherwise ordinary response, as text transmitted to a decoder, who attempts to recover it. The success of this private-message transmission is commonly measured through end-to-end (E2E) recovery, which checks whether the decoder correctly reconstructs the private message. However, E2E recovery cannot distinguish encoding failures from decoding failures or reveal how the private message is represented in the encoder's transmitted text. To address these limitations, we introduce a rolewise decomposition framework that separately examines how private messages are represented in encoder transmitted texts, what decoders recover, and what deterministic verification can establish. Across 7,200 encoder responses, we identify 168 verifiable encodings missed by decoders and 706 responses containing private messages reconstructed through word-length or token-initial patterns not explicitly specified in the encoder's instructions. These findings demonstrate that E2E recovery alone may obscure important distinctions between encoding and decoding, motivating their separate evaluation to better understand and mitigate the safety issues caused by LLM steganography.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Computation and Language (cs.CL); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.30516 — 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 2026-10-24T10:48:23.956Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).