moorcheh/memanto-locomo-results

Memanto — LoCoMo Benchmark Results This dataset contains the full per-question evaluation results of Memanto on the LoCoMo long-term conversational memory benchmark, covering all five stages of a progressive ablation study. It accompanies the paper: Memanto: Typed Semantic Memory with Information-Theoretic Retrieval for Long-Horizon AgentsSeyed Moein Abtahi, Rasa Rahnema, Hetkumar Patel, Neel Patel, Majid Fekri, Tara KhaniMoorcheh AI / EdgeAI Innovations[arXiv preprint] ·… See the full description on the dataset page: https://huggingface.co/datasets/moorcheh/memanto-locomo-results.