Zobrazeno 1 - 10
of 4 108
pro vyhledávání: '"Schedl, A."'
Language models frequently inherit societal biases from their training data. Numerous techniques have been proposed to mitigate these biases during both the pre-training and fine-tuning stages. However, fine-tuning a pre-trained debiased language mod
Externí odkaz:
http://arxiv.org/abs/2409.19541
Most recommender systems adopt collaborative filtering (CF) and provide recommendations based on past collective interactions. Therefore, the performance of CF algorithms degrades when few or no interactions are available, a scenario referred to as c
Externí odkaz:
http://arxiv.org/abs/2409.17864
Autor:
Schedl, Markus, Lesota, Oleg, Brandl, Stefan, Lotfi, Mohammad, Ticona, Gustavo Junior Escobedo, Masoudian, Shahed
Cognitive biases have been studied in psychology, sociology, and behavioral economics for decades. Traditionally, they have been considered a negative human trait that leads to inferior decision-making, reinforcement of stereotypes, or can be exploit
Externí odkaz:
http://arxiv.org/abs/2408.12492
Recent work suggests that music recommender systems are prone to disproportionally frequent recommendations of music from countries more prominently represented in the training data, notably the US. However, it remains unclear to what extent feedback
Externí odkaz:
http://arxiv.org/abs/2408.11565
Autor:
Saeed, Muhammad Saad, Nawaz, Shah, Zaheer, Muhammad Zaigham, Khan, Muhammad Haris, Nandakumar, Karthik, Yousaf, Muhammad Haroon, Sajjad, Hassan, De Schepper, Tom, Schedl, Markus
Multimodal networks have demonstrated remarkable performance improvements over their unimodal counterparts. Existing multimodal networks are designed in a multi-branch fashion that, due to the reliance on fusion strategies, exhibit deteriorated perfo
Externí odkaz:
http://arxiv.org/abs/2408.07445
Autor:
Liaqat, Muhammad Irzam, Nawaz, Shah, Zaheer, Muhammad Zaigham, Saeed, Muhammad Saad, Sajjad, Hassan, De Schepper, Tom, Nandakumar, Karthik, Schedl, Muhammad Haris Khan Markus
Multimodal learning has demonstrated remarkable performance improvements over unimodal architectures. However, multimodal learning methods often exhibit deteriorated performances if one or more modalities are missing. This may be attributed to the co
Externí odkaz:
http://arxiv.org/abs/2407.16243
Segmenting text into sentences plays an early and crucial role in many NLP systems. This is commonly achieved by using rule-based or statistical methods relying on lexical features such as punctuation. Although some recent works no longer exclusively
Externí odkaz:
http://arxiv.org/abs/2406.16678
Autor:
Escobedo, Gustavo, Moscati, Marta, Muellner, Peter, Kopeinik, Simone, Kowald, Dominik, Lex, Elisabeth, Schedl, Markus
Users' interaction or preference data used in recommender systems carry the risk of unintentionally revealing users' private attributes (e.g., gender or race). This risk becomes particularly concerning when the training data contains user preferences
Externí odkaz:
http://arxiv.org/abs/2406.11505
Autor:
Saeed, Muhammad Saad, Nawaz, Shah, Tahir, Muhammad Salman, Das, Rohan Kumar, Zaheer, Muhammad Zaigham, Moscati, Marta, Schedl, Markus, Khan, Muhammad Haris, Nandakumar, Karthik, Yousaf, Muhammad Haroon
The advancements of technology have led to the use of multimodal systems in various real-world applications. Among them, the audio-visual systems are one of the widely used multimodal systems. In the recent years, associating face and voice of a pers
Externí odkaz:
http://arxiv.org/abs/2404.09342
Bias mitigation of Language Models has been the topic of many studies with a recent focus on learning separate modules like adapters for on-demand debiasing. Besides optimizing for a modularized debiased model, it is often critical in practice to con
Externí odkaz:
http://arxiv.org/abs/2401.16457