The paper introduces GLASS, a training‑free framework that personalizes text generation by combining global user‑style priors extracted with sparse autoencoders and local contrastive style vectors built from clustered interaction scenarios. During inference, GLASS injects these global and local vectors into different model layers, achieving context‑aware personalization without retrieval or parameter updates. Experiments on LaMP and LongLaMP datasets demonstrate that GLASS surpasses retrieval, fine‑tuning, and other steering baselines on ROUGE metrics and LLM‑as‑judge evaluations, and its SAE‑based representations show robustness to topic and length shifts, indicating better disentanglement of style from content.
By Liuji Chen, Zeyu Zhang, Xinyuan Zhang, Shuai Nie, Qiang Liu, Shu Wu, Liang Wang
The paper introduces Seq-Drifting, a new framework that enables diffusion language models to generate an entire text sequence in a single forward pass. By applying a drift field to continuous token representations, the model learns a direct mapping from Gaussian noise to text, using attraction to guide outputs toward contextually plausible tokens and repulsion to prevent repetition and collapse. Experiments show that Seq-Drifting achieves a perplexity of 36.01 on LM1B and performs competitively in unconditional, conditional, translation, summarization, and reasoning tasks.
By Jiayu Liu, Ruizhi Zhao, Tianyun Ji, Yee Whye Teh, Zhenya Huang, Ning Miao
arXiv:2510.17881v4 Announce Type: replace
Abstract: Large language models (LLMs) are typically aligned with population-level preferences, despite substantial variation across individual users. We int...
By Yizhuo Chen, Xin Liu, Ruijie Wang, Zheng Li, Pei Chen, Changlong Yu, Qingyu Yin, Priyanka Nigam, Meng Jiang, Bing Yin
arXiv:2610.19166v1 Announce Type: new
Abstract: Meta-evaluation of text generation metrics and LLM-judges is primarily conducted using metric-human correlations. However, prior work has shown that co...
By Anubhav Jangra, Adrian de Wynter
The paper "Evaluating Style-Personalized Text Generation: Challenges and Directions" examines the difficulties of assessing text that is tailored to individual users’ styles. It critiques common metrics such as BLEU, embeddings, and LLM-as-judges, and introduces a style discrimination benchmark covering domain discrimination, authorship attribution, and LLM-generated personalized versus non-personalized discrimination across eight writing tasks. The study finds that ensembles of diverse evaluation metrics outperform single-evaluator approaches and offers guidance for reliable assessment of style-personalized generation.
By Anubhav Jangra, Bahareh Sarrafzadeh, Silviu Cucerzan, Adrian de Wynter, Sujay Kumar Jauhar
arXiv:2610.09795v1 Announce Type: new
Abstract: The demand for personalized LLMs is shifting from style imitation toward content quality. We investigate whether self-distillation can bridge this gap...
By Huayi Lai, Jicheng Yang, Min Yi, Chong Meng
arXiv:2606. 08408v1 Announce Type: cross Abstract: We extend activation steering to diffusion language models (DLMs) and study a novel problem that arose due to the inference mechanism of DLMs: Modifying a text in-place to manifest a different concept.
By Ryandito Diandaru, Ikhlasul Akmal Hanif, Fadli Aulawi Al Ghiffari, Ahmed Elshabrawy, Alham Fikri Aji
arXiv:2602.10216v2 Announce Type: replace
Abstract: A single text prompt passed to a diffusion model yields a wide range of visual outputs determined solely by a stochastic process, leaving users wit...
By Pawe{\l} Skier\'s, Emilia Kaczmarczyk, Tomasz Trzci\'nski, Kamil Deja
RangeSWA is a training‑free method designed to improve length extrapolation in continuous diffusion language models. It combines local sliding‑window attention, remote block summaries, and dynamic content retrieval to maintain local detail while accessing long‑range context. The approach also includes positional remapping, conditioning attention correction, and denoising‑stage‑adaptive window scheduling to mitigate generation degradation, resulting in lower perplexity and fewer repetitions in long‑text generation.
By Guangfu Guo, Xiaoqian Lu, Linsey Pang, Weiran Yao, Haolin Chen, Long Cheng, Kunpeng Liu
arXiv:2606. 14943v1 Announce Type: cross Abstract: Causal Transformers model sequences through an autoregressive factorization of the joint distribution, which enables efficient left-to-right decoding and conditional likelihood computation.
By Yinhan Lu, Eric Elmoznino, L\'eo Gagnon, Sarthak Mittal, Tejas Kasetty, Guillaume Lajoie
arXiv:2602. 21219v2 Announce Type: replace-cross Abstract: Large Language Model (LLM) personalization holds great promise for tailoring responses by leveraging personal context and history.
By Bo Ni, Branislav Kveton, Samyadeep Basu, Subhojyoti Mukherjee, Leyao Wang, Franck Dernoncourt, Sungchul Kim, Seunghyun Yoon, Zichao Wang, Ruiyi Zhang, Puneet Mathur, Jihyung Kil, Jiuxiang Gu, Nedim Lipka, Yu Wang, Ryan A. Rossi, Tyler Derr
The paper introduces Cross-Preference Learning (CPL), a training framework that explicitly models the complementary strengths of sentence-level and context-aware machine translation. By incorporating intra- and cross-condition preferences into the optimization objective, CPL provides targeted supervision to leverage useful contextual signals while remaining robust to uninformative context. Experiments on multiple public context-aware MT tasks with models such as Qwen3-4B, Qwen3-8B, and Llama-3-8B-Instruct show consistent improvements in translation quality and robustness without altering the model architecture.
By Ying Li, Xinglin Lyu, Junhui Li, Jinlong Yang, Hengchao Shang, Min Zhang, Shimin Tao, Daimeng Wei