ALTK10 min read
ALTK-Evolve: How learned memory tops the τ²-bench leaderboard on customer service tasks without a frontier model
ALTK-Evolve learns and applies a playbook using GPT-4.1, outperforming the retail entries on τ²-bench, a customer service agent benchmark—without updating model weights.
By Ashwath Vaithinathan Aravindan, Vatche Isahagian, Vinod Muthusamy, Jayaram Radhakrishnan, Evelyn Duesterwald, Gaodan Fang, Punleuk Oum, G Thomas, Michał Ulewicz, Merve Unuvar, Ayhan SebinHugging Face Blog

Read the full story
Originally published by Hugging Face Blog on Oct 20, 2026.
Comments
Sign in to join the discussion.