Peter Knees
YOU?
Author Swipe
View article: Understanding Verbatim Memorization in LLMs Through Circuit Discovery
Understanding Verbatim Memorization in LLMs Through Circuit Discovery Open
Underlying mechanisms of memorization in LLMs -- the verbatim reproduction of training data -- remain poorly understood. What exact part of the network decides to retrieve a token that we would consider as start of memorization sequence? H…
View article: Nuanced Music Emotion Recognition via a Semi-Supervised Multi-Relational Graph Neural Network
Nuanced Music Emotion Recognition via a Semi-Supervised Multi-Relational Graph Neural Network Open
Music emotion recognition (MER) seeks to understand the complex emotional landscapes elicited by music, acknowledging music’s profound social and psychological roles beyond traditional tasks such as genre classification or content similari…
View article: Enhancing Sequential Music Recommendation with Negative Feedback-informed Contrastive Learning
Enhancing Sequential Music Recommendation with Negative Feedback-informed Contrastive Learning Open
Modern music streaming services are heavily based on recommendation engines to serve content to users. Sequential recommendation -- continuously providing new items within a single session in a contextually coherent manner -- has been an e…
View article: MuRS 2024: 2nd Music Recommender Systems Workshop
MuRS 2024: 2nd Music Recommender Systems Workshop Open
Music recommendation has been relevant to the Recommender Systems (RecSys) community since the early days. With the growth of music streaming platforms, algorithmic recommendations have become critical in the music industry. However, many …
View article: The Need for Tailored Support for Long-Term Students in Informatics Bachelor’s Programmes
The Need for Tailored Support for Long-Term Students in Informatics Bachelor’s Programmes Open
Long-term students in higher education are characterised by extended study durations well beyond the expected or typical completion time. Extended study durations pose a persistent problem to higher education institutions in Austria, as th…
View article: Recommender Systems: Techniques, Effects, and Measures Toward Pluralism and Fairness
Recommender Systems: Techniques, Effects, and Measures Toward Pluralism and Fairness Open
Recommender systems are widely used in various applications, such as online shopping, social media, and news personalization. They can help systems by delivering only the most relevant and promising information to their users and help peop…
View article: Leveraging Negative Signals with Self-Attention for Sequential Music Recommendation
Leveraging Negative Signals with Self-Attention for Sequential Music Recommendation Open
Music streaming services heavily rely on their recommendation engines to continuously provide content to their consumers. Sequential recommendation consequently has seen considerable attention in current literature, where state of the art …
View article: Leveraging Negative Signals with Self-Attention for Sequential Music Recommendation
Leveraging Negative Signals with Self-Attention for Sequential Music Recommendation Open
Music streaming services heavily rely on their recommendation engines to continuously provide content to their consumers. Sequential recommendation consequently has seen considerable attention in current literature, where state of the art …
View article: Digital Humanism: The Time Is Now
Digital Humanism: The Time Is Now Open
Digital humanism highlights the complex relationships between people, society, nature, and machines. It has been embraced by a growing community of individuals and groups who are setting directions that may change current paradigms. Here w…
View article: On the Impact and Interplay of Input Representations and Network Architectures for Automatic Music Tagging
On the Impact and Interplay of Input Representations and Network Architectures for Automatic Music Tagging Open
Automatic music tagging systems have once more gained relevance over the last years, not least through their use in applications such as music recommender systems. State-of-the-art systems are based on a variant of convolutional neural net…
View article: On the Impact and Interplay of Input Representations and Network Architectures for Automatic Music Tagging
On the Impact and Interplay of Input Representations and Network Architectures for Automatic Music Tagging Open
Automatic music tagging systems have once more gained relevance over the last years, not least through their use in applications such as music recommender systems. State-of-the-art systems are based on a variant of convolutional neural net…
View article: A Reproducibility Study on User-centric MIR Research and Why it is Important
A Reproducibility Study on User-centric MIR Research and Why it is Important Open
Reproducibility of results is a central pillar of scientific work. In music information retrieval research, this is widely acknowledged and practiced by the community by re-implementing algorithms and re-validating machine learning experim…
View article: Enabling FAIR use of Ethnomusicology Data – Through Distributed Repositories, Linked Data and Music Information Retrieval
Enabling FAIR use of Ethnomusicology Data – Through Distributed Repositories, Linked Data and Music Information Retrieval Open
Recordings of musical practices are kept in various public institutions and private depositories around the world. They constitute valuable data for ethnomusicological research and are substantial for the world's musical heritage. At the m…
View article: Machine Learning Applied to Music/Audio Signal Processing
Machine Learning Applied to Music/Audio Signal Processing Open
Over the past two decades, the utilization of machine learning in audio and music signal processing has dramatically increased [...]
View article: Scaling Up Broken Systems? Considerations from the Area of Music Streaming
Scaling Up Broken Systems? Considerations from the Area of Music Streaming Open
We discuss the effects and characteristics of disruptive business models driven by technology, exemplified by the developments in music distribution and consumption over the last 20 years. Starting from a historical perspective, we offer i…
View article: Content-driven Music Recommendation: Evolution, State of the Art, and Challenges
Content-driven Music Recommendation: Evolution, State of the Art, and Challenges Open
The music domain is among the most important ones for adopting recommender systems technology. In contrast to most other recommendation domains, which predominantly rely on collaborative filtering (CF) techniques, music recommenders have t…
View article: Unsupervised cross-modal audio representation learning from unstructured multilingual text
Unsupervised cross-modal audio representation learning from unstructured multilingual text Open
We present an approach to unsupervised audio representation learning. Based on a triplet neural network architecture, we harnesses semantically related cross-modal information to estimate audio track-relatedness. By applying Latent Semanti…
View article: Unsupervised Cross-Modal Audio Representation Learning from Unstructured\n Multilingual Text
Unsupervised Cross-Modal Audio Representation Learning from Unstructured\n Multilingual Text Open
We present an approach to unsupervised audio representation learning. Based\non a triplet neural network architecture, we harnesses semantically related\ncross-modal information to estimate audio track-relatedness. By applying Latent\nSema…
View article: Intelligent User Interfaces for Music Discovery
Intelligent User Interfaces for Music Discovery Open
Assisting the user in finding music is one of the original motivations that led to the establishment of Music Information Retrieval (MIR) as a research field. This encompasses classic Information Retrieval inspired access to music reposito…
View article: Proceedings of the 1st Workshop on Human-Centric Music Information Research Systems
Proceedings of the 1st Workshop on Human-Centric Music Information Research Systems Open
Technology and music have a centuries old history of coexistence: from luthiers to music information research. The emergence of machine learning for artificial intelligence in music technology has the potential to change the way music is e…
View article: Multi-Task Learning of Tempo and Beat: Learning One to Improve the Other
Multi-Task Learning of Tempo and Beat: Learning One to Improve the Other Open
In this paper, we propose a multi-task learning approach for simultaneous tempo estimation and beat tracking of musical audio. The system shows state-of-the-art performance for both tasks on a wide range of data, but has another fundamenta…
View article: Intelligent User Interfaces for Music Discovery: The Past 20 Years and What's to Come
Intelligent User Interfaces for Music Discovery: The Past 20 Years and What's to Come Open
Providing means to assist the user in finding music is one of the original motivations underlying the research field known as Music Information Retrieval (MIR). Therefore, already the first edition of ISMIR in the year 2000 called for pape…
View article: Multi-Task Music Representation Learning from Multi-Label Embeddings
Multi-Task Music Representation Learning from Multi-Label Embeddings Open
This paper presents a novel approach to music representation learning. Triplet loss based networks have become popular for representation learning in various multimedia retrieval domains. Yet, one of the most crucial parts of this approach…
View article: Session-Based Hotel Recommendations: Challenges and Future Directions
Session-Based Hotel Recommendations: Challenges and Future Directions Open
In the year 2019, the Recommender Systems Challenge deals with a real-world task from the area of e-tourism for the first time, namely the recommendation of hotels in booking sessions. In this context, this article aims at identifying and …
View article: Towards multi-instrument drum transcription
Towards multi-instrument drum transcription Open
Automatic drum transcription, a subtask of the more general automatic music transcription, deals with extracting drum instrument note onsets from an audio source. Recently, progress in transcription performance has been made using non-nega…
View article: Indicators of Country Similarity in Terms of Music Taste, Cultural, and Socio-economic Factors
Indicators of Country Similarity in Terms of Music Taste, Cultural, and Socio-economic Factors Open
Considering the cultural background of users is known to improve recommender systems for multimedia items. In this work, we focus on music and analyze user demographics and music listening events in a large corpus (120,000 users, 109 event…