Metraipolis › archive › dc_7461cababfce

MIRROR: From Imitation to Internalization in LLM Personalization — arXiv:2610.09795

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09795
License (as archived)
CC0-1.0
Archived
2026-10-08T11:11:28.101Z
SHA-256 of the archived text
c7a9d92cb21b4b7285fd3393ec00b62c2fab7cd854fb530a84dfdcbb3f76ca38
Stable URL of this page
https://metraipolis.com/documento/dc_7461cababfce

How to cite

MIRROR: From Imitation to Internalization in LLM Personalization — arXiv:2610.09795 — arXiv (en), https://arxiv.org/abs/2610.09795 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7461cababfce (Metraipolis archive dc_7461cababfce, sha256 c7a9d92cb21b4b7285fd3393ec00b62c2fab7cd854fb530a84dfdcbb3f76ca38)
BibTeX
@misc{dc7461cababfce,
  title = {MIRROR: From Imitation to Internalization in LLM Personalization — arXiv:2610.09795},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09795},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7461cababfce; sha256 c7a9d92cb21b4b7285fd3393ec00b62c2fab7cd854fb530a84dfdcbb3f76ca38},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09795. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The demand for personalized LLMs is shifting from style imitation toward content quality. We investigate whether self-distillation can bridge this gap in existing fine-tuning paradigm. To address this limitation, we introduce MIRROR(Meta- personalization by Internalizing Reference-Revealed On-policy Reflections), a novel self-distillation framework that shifts LLM personalization from imitation toward preference internalization. First, we replace reference-token imitation with reference-revealed on-policy self-distillation, aligning the model's next-token distributions along its own generation trajectories with those of its reference-conditioned self, thereby internalizing user preferences rather than reproducing reference this http URL , we introduce MIRROR-F, a focal plug-in that augments on-policy distributional alignment with selective supervision over informative reference tokens, thereby strengthening content generation while preserving user-specific expression. Across three personalized generation benchmarks, two model scales, and complementary reference-based and LLM-based evaluations, MIRROR and MIRROR-F achieve leading overall personalization performance and superior text quality, while exhibiting less catastrophic forgetting than SFT-based baselines on three unseen personalized generation tasks. The gains are consistent across model scales and application scenarios, translating to improved performance in LLM personalization tasks.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.09795 — 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 2026-10-08T10:14:06.126Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).