Get Started
Research questionHow can LLM routers personalize model selection from scarce, inconsistent multi-turn user interactions?The same query may call for different models depending on a user's preferences, which can emerge across several interactions. Capturing those preferences is difficult when user-specific examples are sparse and inconsistently represented.
AI
Machine Learning
Natural Language Processing
Latest papersRecent research connected to this question, newest first.GMTRouter: Personalized LLM Router over Multi-turn User InteractionsThe source evaluates GMTRouter for personalized routing across LLMs. It models user–LLM interactions involving users, models, queries, responses, and turns, and examines adaptation to new users from few-shot data without extensive fine-tuning. Reported evidence is experimental, including accuracy, AUC, and few-shot adaptation results.research paper · Sep 2, 2026
Related questions
How can we evaluate LLM reasoning quality beyond final-answer accuracy across deployment contexts?How can multi-turn LLM tutors personalize progressive guidance while preserving answer correctness?How can personalized LLM agents retrieve time-valid memories of persistent and evolving user states?How can multimodal LLM agents satisfy must-have constraints, rank preferences, and abstain when no solution exists?
Home
Topics
Search
Library