Comparing retrieval-augmentation and parameter-efficient fine-tuning for privacy-preserving personalization of large language models

A Salemi, H Zamani - arXiv preprint arXiv:2409.09510, 2024 - arxiv.org
Privacy-preserving methods for personalizing large language models (LLMs) are relatively
under-explored. There are two schools of thought on this topic:(1) generating personalized …

Social skill training with large language models

D Yang, C Ziems, W Held, O Shaikh… - arXiv preprint arXiv …, 2024 - arxiv.org
People rely on social skills like conflict resolution to communicate effectively and to thrive in
both work and personal life. However, practice environments for social skills are typically out …

Longlamp: A benchmark for personalized long-form text generation

I Kumar, S Viswanathan, S Yerra, A Salemi… - arXiv preprint arXiv …, 2024 - arxiv.org
Long-text generation is seemingly ubiquitous in real-world applications of large language
models such as generating an email or writing a review. Despite the fundamental …

Reasoning-Enhanced Self-Training for Long-Form Personalized Text Generation

A Salemi, C Li, M Zhang, Q Mei, W Kong… - arXiv preprint arXiv …, 2025 - arxiv.org
Personalized text generation requires a unique ability of large language models (LLMs) to
learn from context that they often do not encounter during their standard training. One way to …

Personalized Graph-Based Retrieval for Large Language Models

S Au, CJ Dimacali, O Pedirappagari, N Park… - arXiv preprint arXiv …, 2025 - arxiv.org
As large language models (LLMs) evolve, their ability to deliver personalized and context-
aware responses offers transformative potential for improving user experiences. Existing …

Large Language Model-based Role-Playing for Personalized Medical Jargon Extraction

JH Lim, S Kwon, Z Yao, JP Lalor, H Yu - arXiv preprint arXiv:2408.05555, 2024 - arxiv.org
Previous studies reveal that Electronic Health Records (EHR), which have been widely
adopted in the US to allow patients to access their personal medical information, do not …

[PDF][PDF] Controllable Text Generation To Evaluate Linguistic Abilities of Italian LLMs

State-of-the-art Large Language Models (LLMs) demonstrate exceptional proficiency across
diverse tasks, yet systematic evaluations of their linguistic abilities remain limited. This paper …