VitaBench 2.0 introduces a benchmark for long-term personalized and proactive agent behavior, with results indicating substantial gaps in current frontier LLMs.
arXiv preprint arXiv:2409.20296 , year=
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
Low agreeableness massively shifts multi-agent LLM communication yet barely hurts coding milestones, while the same prompt sharply degrades research milestones and collapses bargaining agreements.
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.
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.
MINTEval benchmark shows current memory-augmented systems average 27.9% accuracy on long-horizon interference tasks, limited by retrieval and memory construction with degradation from intervening updates.
citing papers explorer
-
VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactions
VitaBench 2.0 introduces a benchmark for long-term personalized and proactive agent behavior, with results indicating substantial gaps in current frontier LLMs.
-
When Does Personality Composition Matter for Multi-Agent LLM Teams?
Low agreeableness massively shifts multi-agent LLM communication yet barely hurts coding milestones, while the same prompt sharply degrades research milestones and collapses bargaining agreements.
-
Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
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.
-
PAFO: Pareto Fairness Optimization for Personalized Reward Modeling
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.
-
MINTEval: Evaluating Memory under Multi-Target Interference in Long-Horizon Agent Systems
MINTEval benchmark shows current memory-augmented systems average 27.9% accuracy on long-horizon interference tasks, limited by retrieval and memory construction with degradation from intervening updates.