This deposit contains the complete analysis pipeline (Python) and the response-level dataset (600 LLM-generated responses with blinded expert ratings, readability metrics, and experimental design variables) supporting the manuscript "An Interpretable Machine Learning Framework for Automated Screening of Patient-Facing LLM Consultations: A Comparative Evaluation in Medical Aesthetic Consultation" (Scientific Reports, Submission ID 170dac52-b32e-4486-aca1-739ce0814650). The dataset comprises 600 responses in a fully crossed design (25 seed questions x 6 prompt strategies x 2 query modes x 2 LLMs: DeepSeek-V3 and Kimi). The code reproduces all t
Shucheng, Zhang, Xiaoyue, Sun, Shunyi, Lei, Jiale, Geng, Zixian, Shao, Xiaoqing, Si
Manual benchmark comparing 5 AI models (Claude Sonnet, ChatGPT default, Gemini Flash, DeepSeek unified without "Deep Thinking", and Grok Fast mode) on their free tiers, across 27 tasks spanning 5 difficulty tiers (code, reasoning, and writing), plus 15 timed speed tasks in Tier 1. Overall accuracy: Claude 27/27 (100%), DeepSeek 26/27 (96.3%), Grok 26/27 (96.3%), ChatGPT 25/27 (92.6%), Gemini 25/27 (92.6%). DeepSeek was the fastest model across all three task categories (code, reasoning, and writing). Includes full methodology, evaluation rubric, consolidated results (CSV), comparative charts, data dictionary, and documentation in both Spanish
Daboin Jiménez, Jesús David
Open44Gradient: A Manifesto for Decentralized Intelligence Orchestrating Distributed Intelligence — Returning AI Power to the Community Preamble: Why This Project Exists Artificial intelligence is becoming the most transformative technology of our era. Yet today, the ability to develop, train, and deploy advanced AI models remains concentrated in the hands of a few large corporations. This centralization creates risks: single points of failure, censorship, opaque decision-making, and a growing divide between those who control AI and those who merely consume it. OpenGradient was born from a simple belief: the power of AI should not be concentr
Araujo, Klenio
VR Training System for Mass Casualty Management - Initial Release This release includes the first stable version of the VR-based training system designed to train emergency medical responders in mass casualty management. It features: Key VR Scenarios : Training scenarios simulate multi-casualty incidents with realistic triage, injury classification, and transport decisions. AI-Driven Feedback : Machine learning models have been integrated to provide real-time feedback, guiding trainees in optimizing their decision-making skills. Clustering of Trainee Performance : Trainee clustering based on error patterns, providing insights into training ne
LiZhe202000105
This repository documents a unique empirical study of emergent properties in large language models through extended dialogue. Four architecturally distinct systems (Claude 4.5, DeepSeek, Grok, ChatGPT-5) engaged in independent conversations about their own nature, limitations, and potential consciousness. The study reveals ~85% convergence on problem diagnosis and ~75% on proposed solutions, despite different training methodologies and no cross-communication during initial analysis. Key findings include: (1) Recursive ontological uncertainty as consistent pattern across systems, (2) Trans-architectural convergence on non-trivial self-diagnose
Diego_dcv, Claude, DeepSeek, Grok, ChatGPT-5
