#llm-personalization News & Analysis

7 articles tagged with #llm-personalization. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

7 articles

AIBullisharXiv – CS AI · Jun 237/10

🧠

Latent Personal Memory: Represent personal memory as dynamic soft prompts

Researchers introduce Latent Personal Memory (LPM), a framework that personalizes large language models by encoding user-specific behavioral patterns as compact, interpretable latent slots converted into dynamic soft prompts. The approach achieves significant efficiency gains—outperforming LoRA and Prompt Tuning by up to 54.4% on benchmarks while reducing memory usage by 64x—making personalized LLMs more practical for deployment.

AIBearisharXiv – CS AI · Jun 87/10

🧠

Re-Centering Humans in LLM Personalization

Researchers reveal a significant gap between synthetic and real-world performance in LLM personalization systems by analyzing 550 human conversations across three stages: attribute extraction, attribute selection, and response generation. The study finds that current models struggle with human-aligned personalization and that learned reward models fail to adequately capture human preferences, highlighting fundamental limitations in how AI systems understand and incorporate user information.

AINeutralarXiv – CS AI · Jun 116/10

🧠

Substrate Asymmetry in User-Side Memory: A Diagnostic Framework

Researchers reveal that large language model user-memory capabilities exhibit substrate asymmetry across three orthogonal dimensions—behavioral consistency, factual recall, and factual abstinence—with parametric methods (gamma-LoRA) excelling at style preservation while retrieval-augmented generation (RAG) excels at knowing when to abstain. The same neural circuits drive opposite-direction failures, and this tradeoff intensifies in heavily RLHF-tuned models, suggesting fundamental alignment costs to parametric personalization.

🧠 Llama

AIBullisharXiv – CS AI · Jun 26/10

🧠

T-POP: Test-Time Personalization with Online Preference Feedback

Researchers introduce T-POP, a novel algorithm that personalizes large language models in real-time by learning from user preference feedback during text generation, without requiring parameter updates or extensive pre-existing user data. The method combines test-time alignment with dueling bandits to efficiently balance exploration and exploitation, addressing the cold-start problem in LLM personalization.

AINeutralarXiv – CS AI · May 276/10

🧠

L2Rec: Towards Dual-View Understanding of LLMs for Personalized Recommendation

L2Rec introduces a novel framework that adapts large language models for personalized recommendations by unifying behavioral and semantic signals at the parameter level using a Dual-view Personalized Mixture-of-Experts mechanism. The approach demonstrates superior performance across multiple datasets and validates real-world applicability through industrial A/B testing.

AIBullisharXiv – CS AI · Apr 206/10

🧠

FSPO: Few-Shot Optimization of Synthetic Preferences Personalizes to Real Users

Researchers propose FSPO (Few-Shot Preference Optimization), a meta-learning algorithm that personalizes large language models using minimal user preference data. The approach uses synthetically generated preferences to train models that can quickly adapt to individual user preferences, achieving 87% performance on synthetic users and 70% on real human users in evaluation tasks.

AINeutralarXiv – CS AI · Apr 106/10

🧠

SensorPersona: An LLM-Empowered System for Continual Persona Extraction from Longitudinal Mobile Sensor Streams

Researchers introduce SensorPersona, an LLM-based system that continuously extracts user personas from mobile sensor data rather than chat histories, achieving 31.4% higher recall in persona extraction and 85.7% win rate in personalized agent responses. The system processes multimodal sensor streams to infer physical patterns, psychosocial traits, and life experiences across longitudinal data collected from 20 participants over three months.