خانه / اخبار / Hugging Face Blog opensource ALTK-Evolve: How learned memory tops the τ²-bench leaderboard on customer service tasks without a frontier model متن کامل در منبع اشتراک بیشتر بخوانید اخبار مرتبط همه اخبار Hugging Face Blog Token Pooling-Aware Training for Multi-Vector Embedding Models 3 روز پیش → Hugging Face Blog Decision Models in Hugging Face 4 روز پیش → Hugging Face Blog Dharma Field Extractor: Any Document In, Structured Fields in JSON Out 2026/11/04 →