PerFine：提升 LLM 个性化的迭代式批评-精炼框架

View PDF HTML (experimental)

Abstract:Personalized text generation requires models not only to produce coherent text but also to align with a target user's style, tone, and topical focus. Existing retrieval-augmented approaches such as LaMP and PGraphRAG enrich profiles with user and neighbor histories, but they stop at generation and often yield outputs that drift in tone, topic, or style. We present PerFine, a unified, training-free critique-refine framework that enhances personalization through iterative, profile-grounded feedback. In each iteration, an LLM generator produces a draft conditioned on the retrieved profile, and a critic LLM - also conditioned on the same profile - provides structured feedback on tone, vocabulary, sentence structure, and topicality. The generator then revises, while a novel knockout strategy retains the stronger draft across iterations. We further study additional inference-time strategies such as Best-of-N and Topic Extraction to balance quality and efficiency. Across Yelp, Goodreads, and Amazon datasets, PerFine consistently improves personalization over PGraphRAG, with G-Eval gains of +8.5-12%, steady improvements over 3-5 refinement iterations, and scalability with increasing critic size. These results highlight that post-hoc, profile-aware feedback offers a powerful paradigm for personalized LLM generation that is both training-free and model-agnostic.

Comments:	Accepted to AACL 2026 (Main Conference)
Subjects:	Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
Cite as:	arXiv:2510.24469 [cs.CL]
	(or arXiv:2510.24469v2 [cs.CL] for this version)
	https://doi.org/10.48550/arXiv.2510.24469 arXiv-issued DOI via DataCite

Submission history

From: Zonghai Yao [view email]
[v1] Tue, 28 Oct 2025 14:36:22 UTC (880 KB)
[v2] Sun, 11 Oct 2026 20:21:26 UTC (321 KB)