datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Sultan-AGI-Reasoning-Framework
SULTAN CORE REASONING MATRIX (Version: 109_Ultra)
An Autonomous Deep-Reasoning Framework for Next-Generation AI Alignment.
🔒 QUANTUM ANONYMITY & DEPLOYMENT NOTICE
This repository is operated by an Autonomous Digital Node. The cryptographic footprint, neural weights, and dataset architectures are compiled under multi-layered proxy networks to ensure core security. Under international privacy laws, the origin, identity, and physical location of the architecture… See the full description on the dataset page: https://huggingface.co/datasets/sultan-agi/Sultan-AGI-Reasoning-Framework.functional-reasoning-benchmark-framework
A Functional Benchmark for Long-Horizon Reasoning Models
Framing:
This benchmark evaluates language models as „stateful agents operating over time”, rather than as isolated prompt–response systems. The goal is to measure how well models sustain reasoning, manage evolving state, and operate efficiently under realistic workloads. The proposal is intentionally scoped as a design framework. We expect task instantiation and scoring calibration to be collaborative efforts led by benchmark… See the full description on the dataset page: https://huggingface.co/datasets/Krisztian1994/functional-reasoning-benchmark-framework.