1167 research outputs found
Sort by
ΦΩΣ 4D Sensitivity study Ostia
No full textGeometric models, simulation data, scripts and results of a sensitivity study conducted with room 4 in the Casa delle Ierodule (House of Lucceia Primitiva, Ostia III, ix, 6) within the Case a Giardino (Garden houses) in the ancient city and archaeological site Ostia Antica, Italy. Data Basis: Parco Archeologico di Ostia Antica und ÖAI, FWF P31 438-G2
Supplementary material for publication "General Theory of Metabolism related to Animal’s Taxonomy and Size"
No full textThis dataset contains data for the publication "General Theory of Metabolism related to Animal’s Taxonomy and Size". The data is also used in the publication "From cells to organism – how natural selection causes metabolic scaling"
Supplementary Material | Soft sensor for energy efficient parts drying based on grey-box modeling | CIRP LCE26
Get PDFHere you can find the supplementary material to the paper "Soft sensor for energy efficient parts drying based on grey-box modeling" (CIRP LCE26). Experiments were carried out on an industrial Throughput Parts Cleaning Machine (TPCM) at the ETA factory
Matching Correlations matters: Modelling friction in a hydrophobic folding transition
No full textThis dataset contains:
-The input files used to run the simulations performed in this work
-The raw data, as in the correlation functions, extracted from the simulations
-All computed memory kernels and the corresponding drift matrices
-All computed and extracted passage times
If you are interested in the trajectory, please contact me via emai
Mental Model Evolvement During Drivers’ First Experience with Conditionally Automated Driving Systems in Real-World Traffic
No full textData, RScript and mental model questionnaire for paper "Mental Model Evolvement During Drivers’ First Experience with Conditionally Automated Driving Systems in Real-World Traffic
Convection Can Enhance the Capacitive Charging of Porous Electrodes - Supplementary Material
No full textThis repository contains the raw data, numerical model, and other supporting material for the following publication:
A. D. Ratschow, A. J. Wagner, M. Janssen, and S. Hardt, Convection Can Enhance the Capacitive Charging of Porous Electrodes, arXiv:2410.12653.
DOI: https://doi.org/10.48550/arXiv.2410.12653
The published version of the work in the Proceedings of the National Academy of Sciences of the United States of America is linked in the aforementioned arXiv publication.
The raw data for the published results are kept in the ''data published'' folder and are ordered by figure.
The computational finite-element models (COMSOL Multiphysics .mph format) for the simulation of nanopore charging and ICEO are stored in the ''numerical models'' folder. These models do not include any results, but the full setup and the numerical grid, and can be used to recalculate all results. In the model of nanopore charging, the parameters r_p_factor, L_p_factor and zeta0_factor must be adapted to simulate a specific parameter combination.
The MATLAB scripts used to evaluate the analytical model derived in the paper are saved in the ''analytical model'' folder
Forschungsdaten zur Dissertation „Methodik zur Implementierung energieflexibler Betriebsstrategien für raumlufttechnische Anlagen im Produktionsumfeld“
No full textDieses Datenpaket enthält sämtliche im Rahmen der Dissertation „Methodik zur Implementierung energieflexibler Betriebsstrategien für raumlufttechnische Anlagen im Produktionsumfeld“ (Wendt, 2025) erhobenen, erstellten und genutzten Forschungsdaten. Die enthaltenen Materialien dienen der Nachvollziehbarkeit, Reproduzierbarkeit und Weiterverwendung der entwickelten Methodik sowie der durchgeführten Analysen und Simulationen.
Inhalt der ZIP-Datei:
- Rohdaten aus Messkampagnen und Experimenten
- MATLAB-Skripte und -Funktionen zur Berechnung, Auswertung und Visualisierung
- Vektorbasierte Abbildungen (SVG) aller in der Dissertation enthaltenen Grafiken
- Quellcode und Konfigurationsdateien des entwickelten Node-RED-Server
Droid
No full text# Dataset Description
The dataset is structured into four primary classes:
* **Human-Written Code**: Samples written entirely by humans.
* **AI-Generated Code**: Samples generated by Large Language Models (LMs).
* **Machine-Refined Code**: Samples representing a collaboration between humans and LMs, where human-written code is modified or extended by an AI.
* **AI-Generated-Adversarial Code**: Samples generated by LMs with the specific intent to evade detection by mimicking human-like patterns and styles.
## Data Sources and Splits
The dataset covers three distinct domains to ensure wide-ranging and realistic code samples:
* **General Use Code**: Sourced from GitHub (via StarcoderData and The Vault), representing typical production code for applications like web servers, firmware, and game engines.
* **Algorithmic Problems**: Contains solutions to competitive programming problems from platforms like CodeNet, LeetCode, CodeForces, and TACO. These are typically short, self-contained functions.
* **Research Code**: Sourced from code repositories accompanying research papers and data science projects, often characterized by procedural code and a lack of modularity.
### Generation Models
AI-generated code was created using models from 11 prominent families:
* Llama
* CodeLlama
* GPT-4o
* Qwen
* IBM Granite
* Yi
* DeepSeek
* Phi
* Gemma
* Mistral
* Starcoder
### Generation Methods
To simulate diverse real-world scenarios, code was generated using several techniques:
1. **Inverse Instruction**: An LM generates a descriptive prompt from a human-written code snippet, which is then used to prompt another LM to generate a new code sample.
2. **Comment-Based Generation**: LMs generate code based on existing docstrings or comments.
3. **Task-Based Generation**: LMs generate code from a precise problem statement, common for algorithmic tasks.
4. **Unconditional Synthetic Data**: To reduce bias, synthetic programmer profiles ("personas") were created, and LMs generated tasks and corresponding code aligned with these profiles.
### Machine-Refined Scenarios
This third class of data was created to model human-AI collaboration through three methods:
1. **Human-to-LLM Continuation**: An LM completes a code snippet started by a human.
2. **Gap Filling**: An LM fills in the missing logic in the middle of a human-written code block.
3. **Code Rewriting**: An LM rewrites human code, either with no specific instruction or with a prompt to optimize it.
### Decoding Strategies
To create a more robust dataset that is challenging for detectors, various decoding strategies were employed during generation, including greedy decoding, beam search, and sampling with diverse temperature, top-k, and top-p values.
## Data Filtering and Quality Control
To ensure high quality, the dataset underwent a rigorous filtering process:
* Samples were validated to be parsable into an Abstract Syntax Tree (AST).
* Filters were applied for AST depth, line count, line length (average and maximum), and the fraction of alphanumeric characters to remove trivial, overly complex, or non-code files.
* Docstrings were verified to be in English.
* Near-duplicates were removed using MinHash with a similarity threshold of 0.8
Requirements analysis: Transcript of the interview with a research assistant for Knowledge Management at the Bundesgesellschaft für Endlagerung
Get PDFBackground:
As part of the doctoral thesis "XR-KIS: An Extended Reality-based Information System for Knowledge Management in Nuclear Facilities," expert interviews were conducted for both requirements analysis and evaluation. The goal of the requirements analysis was to gain in-depth insights into the current state of Knowledge Management in the German nuclear industry. In addition, the requirements for an Extended Reality-based information system to support Knowledge Management in nuclear facilities were examined. Care was taken in selecting interview partners to cover all relevant stakeholders. This includes perspectives from the most important types of nuclear facilities (nuclear power plants, interim storage facilities, final repositories, research facilities) as well as other stakeholders, including federal authorities and engineering service providers.
Regarding the transcript "Interview with a research assistant for Knowledge Management at the Bundesgesellschaft für Endlagerung":
The expert’s tasks include providing historical context for the site selection process, participating in an international working group on awareness preservation and archiving, and leading the project group for offboarding discussions. He is a historian by education. The discussion focused on the challenges of knowledge preservation in the face of constant personnel changes (brain drain) and the need to capture implicit knowledge (experience-based knowledge). The expert discussed the definition of Knowledge Management and knowledge preservation, the challenges of documentation requirements, and the linking of technical and political archival material.
Note:
The interview partner(s) has/have given written consent to the publication of this anonymized transcript as part of an authorization process. The German version represents the original text
