Marian Lambert
Marian Lambert has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Thomas Schuster
Fakultät für Wirtschaft und Recht, Wirtschaftsinformatik, Hochschule Pforzheim, Pforzheim, Deutschland
11 shared publications - Britta Stammler
Center for Neurology, Division of Neuropsychology, Hertie-Institute for Clinical Brain Research, Tübingen, Germany
3 shared publications - Hans-Otto Karnath
Center of Neurology, Division of Neuropsychology, Hertie-Institute for Clinical Brain Research, University of Tübingen, Tübingen, Germany
3 shared publications - Kathrin Flammer
Flammer & Gläser UXplain GbR
3 shared publications - Nico Döring
XPACE GmbH
3 shared publications - Lukas Waidelich
Pforzheim University
2 shared publications
Latest Publications
- 10 October 2026
- Published by Springer Nature
- p. 246-260
Sovereign Multi-Agent Collaboration: A Compliance-Aware Guardrail Architecture for Resilient B2B Networks
- 10 October 2026
- Published by Springer Nature
- p. 246-260
Sovereign Multi-Agent Collaboration: A Compliance-Aware Guardrail Architecture for Resilient B2B Networks
- 20 October 2025
- Published by Springer Nature
- p. 104-118
Bridging the Gap: Identifying AI-Suitable Problems and Matching with Solution Providers
- 20 October 2025
- Published by Springer Nature
- p. 104-118
Bridging the Gap: Identifying AI-Suitable Problems and Matching with Solution Providers
- 22 June 2025
- Published by Springer Nature
- p. 254-266
Needle-in-the-Haystack Testing LLMs with a Complex Reasoning Task
- 22 June 2025
- Published by Springer Nature
- p. 254-266
Needle-in-the-Haystack Testing LLMs with a Complex Reasoning Task
- Open Access
Datenschutzkonformer LLM-Einsatz: Eine Open-Source-Referenzarchitektur
The development of Large Language Models (LLMs) has led to significant advancements in natural language processing and enabled numerous applications across various industries. However, many LLM-based solutions operate as open systems relying on cloud services, which pose risks to data confidentiality and security. To address these challenges, organizations require closed LLM systems that comply with data protection regulations while maintaining high performance. In this paper, we present a reference architecture for developing closed, LLM-based systems using open-source technologies. The architecture provides a flexible and transparent solution that meets strict data privacy and security requirements. We analyze the key challenges in implementing such systems, including computing resources, data management, scalability, and security risks. Additionally, we introduce an evaluation pipeline that enables a systematic assessment of system performance and compliance.
Open AccessDatenschutzkonformer LLM-Einsatz: Eine Open-Source-Referenzarchitektur
The development of Large Language Models (LLMs) has led to significant advancements in natural language processing and enabled numerous applications across various industries. However, many LLM-based solutions operate as open systems relying on cloud services, which pose risks to data confidentiality and security. To address these challenges, organizations require closed LLM systems that comply with data protection regulations while maintaining high performance. In this paper, we present a reference architecture for developing closed, LLM-based systems using open-source technologies. The architecture provides a flexible and transparent solution that meets strict data privacy and security requirements. We analyze the key challenges in implementing such systems, including computing resources, data management, scalability, and security risks. Additionally, we introduce an evaluation pipeline that enables a systematic assessment of system performance and compliance.
- 12 September 2024
- Published by Springer Nature
- p. 189-206
Evaluating Privacy Patterns Within Collaborative Frameworks for AI Ecosystem Development
- 12 September 2024
- Published by Springer Nature
- p. 189-206
Evaluating Privacy Patterns Within Collaborative Frameworks for AI Ecosystem Development
- 15 December 2023
- Published by Springer Nature
- p. 158-170
Robustness Analysis of Machine Learning Models Using Domain-Specific Test Data Perturbation
- 15 December 2023
- Published by Springer Nature
- p. 158-170
Robustness Analysis of Machine Learning Models Using Domain-Specific Test Data Perturbation
- 21 September 2023
- Published by Springer Nature
- p. 1-15
Using Large Language Models for the Enforcement of Consumer Rights in Germany
- 21 September 2023
- Published by Springer Nature
- p. 1-15
Using Large Language Models for the Enforcement of Consumer Rights in Germany
- Open Access
Spatial neglect therapy with the augmented reality app 'Negami' for active exploration training; A randomized controlled trial
Background A widely applied and effective rehabilitation method in stroke patients suffering from spatial neglect is the ‘visual exploration training’. Patients improve their ipsilesional bias of attention and orientation by training of exploration movements and search strategies towards the contralesional side of space. Here we investigate the effectiveness of the augmented reality (AR)-based app “Negami” for the treatment of spatial neglect in a randomized control trial. Negami combines a visual exploration training with active, contralesionally oriented rotation of eyes, head, and trunk. Methods Twenty patients with spatial neglect were randomly assigned to the experimental Negami group or to a group receiving standard neglect therapy. Over a period of two weeks, both groups received five training sessions per week (à 25 minutes). Neglect behavior was assessed weekly over a five-week period, with the Negami therapy group receiving a second follow-up assessment at one-to-two-month intervals after completion of training. Results Both groups improved significantly. While the Negami therapy group improved in four of five neglect tests used, the standard therapy group improved in only one of these tests. We observed significantly better improvement in the Negami group already after the first week of training. This difference was also significant after the end of the training as well as one week after the end of training and remained stable one to two months after the end of treatment. Conclusion Negami can be used as an effective alternative or addition to current standard neglect therapy, and may even be superior to it. Trial Registration at “OSF Preregistration”, Registration DOI: https://doi.org/10.17605/OSF.IO/H6XN5 and at WHO approved public trial registry “German Clinical Trials Register”, DRKS-ID: DRKS00031446
Open AccessSpatial neglect therapy with the augmented reality app 'Negami' for active exploration training; A randomized controlled trial
Background A widely applied and effective rehabilitation method in stroke patients suffering from spatial neglect is the ‘visual exploration training’. Patients improve their ipsilesional bias of attention and orientation by training of exploration movements and search strategies towards the contralesional side of space. Here we investigate the effectiveness of the augmented reality (AR)-based app “Negami” for the treatment of spatial neglect in a randomized control trial. Negami combines a visual exploration training with active, contralesionally oriented rotation of eyes, head, and trunk. Methods Twenty patients with spatial neglect were randomly assigned to the experimental Negami group or to a group receiving standard neglect therapy. Over a period of two weeks, both groups received five training sessions per week (à 25 minutes). Neglect behavior was assessed weekly over a five-week period, with the Negami therapy group receiving a second follow-up assessment at one-to-two-month intervals after completion of training. Results Both groups improved significantly. While the Negami therapy group improved in four of five neglect tests used, the standard therapy group improved in only one of these tests. We observed significantly better improvement in the Negami group already after the first week of training. This difference was also significant after the end of the training as well as one week after the end of training and remained stable one to two months after the end of treatment. Conclusion Negami can be used as an effective alternative or addition to current standard neglect therapy, and may even be superior to it. Trial Registration at “OSF Preregistration”, Registration DOI: https://doi.org/10.17605/OSF.IO/H6XN5 and at WHO approved public trial registry “German Clinical Trials Register”, DRKS-ID: DRKS00031446
- Open Access
- 27 February 2023
- Published by JMIR Publications Inc. in JMIR Serious Games
- Vol. 11 (1) , e40651
Negami: An Augmented Reality App for the Treatment of Spatial Neglect After Stroke
Background: A widely applied and effective rehabilitation method for patients experiencing spatial neglect after a stroke is “visual exploration training.” Patients improve their ipsilesional bias of attention and orientation by training exploration movements and search strategies toward the contralesional side of space. In this context, gamification can have a positive influence on motivation for treatment and thus on the success of treatment. In contrast to virtual reality applications, treatment enhancements through augmented reality (AR) have not yet been investigated, although they offer some advantages over virtual reality. Objective: This study aimed to develop an AR-based app (Negami) for the treatment of spatial neglect that combines visual exploration training with active, contralesionally oriented rotation of the eyes, head, and trunk. Methods: The app inserts a virtual element (origami bird) into the real space surrounding the patient, which the patient explores with the camera of a tablet. Subjective reports from healthy elderly participants (n=10) and patients with spatial neglect after stroke (n=10) who trained with the new Negami app were analyzed. Usability, side effects, and game experience were assessed by various questionnaires. Results: Training at the highest defined difficulty level was perceived as differently challenging but not as frustrating by the group of healthy elderly participants. The app was rated with high usability, hardly any side effects, high motivation, and entertainment. The group of patients with spatial neglect after stroke consistently evaluated the app positively on the dimensions of motivation, satisfaction, and fun. Conclusions: The Negami app represents a promising extension by adding AR to traditional exploration training for spatial neglect. Through participants’ natural interaction with the physical surrounding environment during playful tasks, side effects as symptoms of cybersickness are minimized and patients’ motivation appeared to markedly increase. The use of AR in cognitive rehabilitation programs and the treatment of spatial neglect seems promising and should receive further investigation.
- 27 February 2023
- Published by JMIR Publications Inc. in JMIR Serious Games
- Vol. 11 (1) , e40651
Open AccessNegami: An Augmented Reality App for the Treatment of Spatial Neglect After Stroke
Background: A widely applied and effective rehabilitation method for patients experiencing spatial neglect after a stroke is “visual exploration training.” Patients improve their ipsilesional bias of attention and orientation by training exploration movements and search strategies toward the contralesional side of space. In this context, gamification can have a positive influence on motivation for treatment and thus on the success of treatment. In contrast to virtual reality applications, treatment enhancements through augmented reality (AR) have not yet been investigated, although they offer some advantages over virtual reality. Objective: This study aimed to develop an AR-based app (Negami) for the treatment of spatial neglect that combines visual exploration training with active, contralesionally oriented rotation of the eyes, head, and trunk. Methods: The app inserts a virtual element (origami bird) into the real space surrounding the patient, which the patient explores with the camera of a tablet. Subjective reports from healthy elderly participants (n=10) and patients with spatial neglect after stroke (n=10) who trained with the new Negami app were analyzed. Usability, side effects, and game experience were assessed by various questionnaires. Results: Training at the highest defined difficulty level was perceived as differently challenging but not as frustrating by the group of healthy elderly participants. The app was rated with high usability, hardly any side effects, high motivation, and entertainment. The group of patients with spatial neglect after stroke consistently evaluated the app positively on the dimensions of motivation, satisfaction, and fun. Conclusions: The Negami app represents a promising extension by adding AR to traditional exploration training for spatial neglect. Through participants’ natural interaction with the physical surrounding environment during playful tasks, side effects as symptoms of cybersickness are minimized and patients’ motivation appeared to markedly increase. The use of AR in cognitive rehabilitation programs and the treatment of spatial neglect seems promising and should receive further investigation.