{"total":14,"items":[{"citing_arxiv_id":"2606.09038","ref_index":167,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs","primary_cat":"cs.AI","submitted_at":"2026-06-08T05:10:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A survey that maps safety risks in personalized LLMs, introduces a unified taxonomy, and highlights three structural inadequacies in existing research on user-invariant safety, isolated techniques, and short-term evaluations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.07988","ref_index":26,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"PAFO: Pareto Fairness Optimization for Personalized Reward Modeling","primary_cat":"cs.AI","submitted_at":"2026-06-06T05:35:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"PAFO applies Pareto fairness optimization and group-specialized distillation to produce a single personalized reward model that improves accuracy for both majority and minority preference groups without requiring group labels at inference.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02300","ref_index":35,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Beyond Isolated Behaviors: Hierarchical User Modeling for LLM Personalization","primary_cat":"cs.CL","submitted_at":"2026-06-01T14:23:17+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"PHF applies Bourdieu's Theory of Practice to create hierarchical user models for LLM personalization and reports consistent gains on the LaMP benchmark.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00728","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"From Empathy to Personalized Empathy: Adapting Empathetic Strategies to Individual Users","primary_cat":"cs.CL","submitted_at":"2026-05-30T13:49:01+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Introduces personalized empathy task, PersonaEmp dataset from long-term interactions, and PereGRM reward framework that combines empathy evaluation with dynamic criteria for improved adaptation to user personas.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.29372","ref_index":32,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"On the Road to Personalized Code Intelligence: Portraiting and Assisting Developers Based on Their In-IDE Behaviors","primary_cat":"cs.SE","submitted_at":"2026-05-28T05:17:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"VirtualME is a new infrastructure that continuously extracts and interprets in-IDE developer behaviors to build personalized personas, delivering 33.8% better performance on repository-level knowledge Q&A than generic baselines.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.26969","ref_index":12,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Recon: Reconstruction-Guided Reasoning Synthesis for User Modeling","primary_cat":"cs.CL","submitted_at":"2026-05-26T12:55:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Recon scores reasoning traces via action reconstruction fidelity, achieving 54.7% win rate over post-hoc baselines and up to 70% when used to train synthesis models across four domains.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.26072","ref_index":30,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Active Query Synthesis for Preference Learning","primary_cat":"cs.LG","submitted_at":"2026-05-25T17:37:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Info-Synth synthesizes optimal preference queries via mutual information maximization in continuous space and a confidence-aware response model, with extensions for finite pools.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07162","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"CLIPer: Tailoring Diverse User Preference via Classifier-Guided Inference-Time Personalization","primary_cat":"cs.CL","submitted_at":"2026-05-08T02:47:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"CLIPer uses classifier guidance during inference to personalize LLM generations across single and multi-dimensional user preferences without extensive fine-tuning.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.21571","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Separable Expert Architecture: Toward Privacy-Preserving LLM Personalization via Composable Adapters and Deletable User Proxies","primary_cat":"cs.AI","submitted_at":"2026-04-23T11:51:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A separable expert architecture uses base models, LoRA adapters, and deletable per-user proxies to enable privacy-preserving personalization and deterministic unlearning in LLMs.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"is compatible with differentially private stochastic gra- dient descent (DP-SGD) for privacy-preserving shared model improvement. 1 Introduction As LLM personalization becomes widely used, a growing body of work has demonstrated that user preferences can be captured through retrieval-augmented profiles [1], post-hoc parameter merging [2], and personalized reward learning [3, 4]. While some of these approaches oper- ate at the prompt level (e.g., retrieval-augmented pro- files), many encode user-specific information into model weightsθvia fine-tuning, producing models whose pa- rameters entangle contributions from many users. When a user later requests deletion it is unclear how one can remove their data from a model whose weights have"},{"citing_arxiv_id":"2604.11259","ref_index":13,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Mobile GUI Agent Privacy Personalization with Trajectory Induced Preference Optimization","primary_cat":"cs.AI","submitted_at":"2026-04-13T10:12:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TIPO applies preference-intensity weighting and padding gating to stabilize preference optimization for privacy personalization in mobile GUI agents, yielding higher alignment and distinction metrics than prior methods.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"2 Personalization and User Modeling Personalized alignment aims to move beyond population-level be- havior and adapt models to users' preferences, histories, and deci- sion styles [5, 17, 34]. Existing work studies this problem through personalized preference learning, progressive adaptation, and bench- mark construction. Representative examples include P-RLHF [13], which introduces personalized preference learning, PROPER [41], which formulates personalization as progressive refinement, and re- cent benchmarks such as PersonaLens [43] and Persona2Web [10], Mobile GUI Agent Privacy Personalization with Trajectory Induced Preference Optimization Privacy-first Utility-firstObject Task: Open a video link in Chrome"},{"citing_arxiv_id":"2604.09876","ref_index":52,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Efficient Personalization of Generative User Interfaces","primary_cat":"cs.LG","submitted_at":"2026-04-10T20:05:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A dataset revealing high inter-designer disagreement on UI preferences motivates a sample-efficient method that personalizes generative interfaces by embedding new users in the space of prior designers, outperforming baselines in both modeling and user preference.","context_count":1,"top_context_role":"background","top_context_polarity":"support","context_text":"2025. Mllm as a ui judge: Benchmarking multimodal llms for predicting human perception of user interfaces.arXiv preprint arXiv:2510.08783 (2025). [51] Yuhang Ma, Xiaoshi Wu, Keqiang Sun, and Hongsheng Li. 2025. Hpsv3: To- wards wide-spectrum human preference score. InProceedings of the IEEE/CVF International Conference on Computer Vision. 15086-15095. [52] Wendy E. Mackay. 1991. Triggers and Barriers to Customizing Software. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '91). Association for Computing Machinery, New York, NY, USA, 153-160. doi:10.1145/108844.108867 [53] Anne-Sofie Maerten, Li-Wei Chen, Stefanie De Winter, Christophe Bossens, and Johan Wagemans. 2025."},{"citing_arxiv_id":"2604.07629","ref_index":54,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Behavior Latticing: Inferring User Motivations from Unstructured Interactions","primary_cat":"cs.HC","submitted_at":"2026-04-08T22:08:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Behavior latticing synthesizes connections across unstructured user interactions to generate insights into underlying motivations, yielding deeper and more accurate user understanding than task-only models.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Lost in the middle: How language models use long contexts.Transactions of the Association for Computational Linguistics 12 (2024), 157-173. [53] Yibo Lyu, Gongwei Chen, Rui Shao, Weili Guan, and Liqiang Nie. 2026. Per- sonalAlign: Hierarchical Implicit Intent Alignment for Personalized GUI Agent with Long-Term User-Centric Records.arXiv preprint arXiv:2601.09636(2026). [54] Jiaju Ma, Lei Shi, Kenneth Aleksander Robertsen, and Peggy Chi. 2025. Am- bigChat: Interactive Hierarchical Clarification for Ambiguous Open-Domain Question Answering. InProceedings of the 38th Annual ACM Symposium on User Interface Software and Technology. 1-18. [55] Pattie Maes. 1994. Agents that reduce work and information overload.Commun. ACM37, 7 (1994), 30-40."},{"citing_arxiv_id":"2510.17881","ref_index":27,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"POPI: Personalizing LLMs via Optimized Natural Language Preference Inference","primary_cat":"cs.CL","submitted_at":"2025-10-17T23:07:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"POPI distills user preferences into reusable natural-language summaries via a shared inference model and conditions a generator on them, trained jointly with RL to improve personalization quality while cutting context length by up to 10x on benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2412.08812","ref_index":42,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Test-Time Alignment via Hypothesis Reweighting","primary_cat":"cs.LG","submitted_at":"2024-12-11T23:02:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"HyRe personalizes reward models at test time by reweighting an ensemble of heads trained on aggregate preferences, using few target examples to outperform uniform averaging and prior methods on RewardBench and 32 tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}