arXiv:2608. 09507v1 Announce Type: cross Abstract: Natural language user preferences provide an interpretable interface for LLM personalization.
By Yuting Liu, Wei Wu, Jianzhe Zhao, Guibing Guo
arXiv:2609.00251v1 Announce Type: new
Abstract: As people increasingly interact with LLM assistants in daily life, continually adapting to individual preferences has become essential for effective lo...
By EunJeong Hwang, Kushan Mitra, Dan Zhang, Hannah Kim, Estevam Hruschka
HyperTrace is a training‑free framework that personalizes large language models by tracing latent user preferences online. It maintains interpretable natural‑language hypotheses about short‑term intent and long‑term preferences, updating them with an SMC‑style reweighting process driven by an LLM‑based surrogate choice model. Experiments on PRISM and PersonaMem‑v2 demonstrate that HyperTrace improves response alignment, preference prediction, and profile consistency compared to strong online baselines.
By Jianzhi Shen, Keyu Mao, Minghao Shao, Chuanyang Jin, Yusong Wang, Ailiang Lin, Kotaro Funakoshi, Manabu Okumura, Tianmin Shu, Muhammad Shafique
arXiv:2606. 06614v1 Announce Type: cross Abstract: Despite growing interest, most evaluations of large language models' (LLMs') personalization abilities have relied on synthetic data.
By Lechen Zhang, Jiarui Liu, Tal August
The paper introduces an action‑on‑item schema that pairs interaction roles with content embeddings, enabling a shared update mechanism across different user history types such as movies, news, and dialogue. It demonstrates theoretical properties like invariance to relabeling and bounded state changes, and presents the Multi‑Timescale State Hypothesis (MTSH) implemented in PerTIDE. Experiments on PENS, MovieLens, and MIND datasets show that a frozen source‑trained core outperforms random baselines and that PerTIDE achieves significant MRR gains over comparable models.
By Parthiv Chatterjee, Kashish Kanjaria, Vashisth Purani, Sourish Dasgupta, Tanmoy Chakraborty
arXiv:2510. 05342v2 Announce Type: replace-cross Abstract: Direct Preference Optimization (DPO) has emerged as a simple and effective method for aligning large language models.
By Hyung Gyu Rho