Zobrazeno 1 - 10
of 4 087
pro vyhledávání: '"Niehues A"'
Multimodal foundation models aim to create a unified representation space that abstracts away from surface features like language syntax or modality differences. To investigate this, we study the internal representations of three recent models, analy
Externí odkaz:
http://arxiv.org/abs/2411.17666
The small-signal stability of power grids is a well-studied topic. In this work, we give new sufficient conditions for highly heterogeneous mixes of grid-forming inverters (and other machines) that implement a $V$-$q$ droop to stabilize viable operat
Externí odkaz:
http://arxiv.org/abs/2411.10832
In adaptive dynamical networks, the dynamics of the nodes and the edges influence each other. We show that we can treat such systems as a closed feedback loop between edge and node dynamics. Using recent advances on the stability of feedback systems
Externí odkaz:
http://arxiv.org/abs/2411.10387
Autor:
Ahmad, Ibrahim Said, Anastasopoulos, Antonios, Bojar, Ondřej, Borg, Claudia, Carpuat, Marine, Cattoni, Roldano, Cettolo, Mauro, Chen, William, Dong, Qianqian, Federico, Marcello, Haddow, Barry, Javorský, Dávid, Krubiński, Mateusz, Lam, Tsz Kin, Ma, Xutai, Mathur, Prashant, Matusov, Evgeny, Maurya, Chandresh, McCrae, John, Murray, Kenton, Nakamura, Satoshi, Negri, Matteo, Niehues, Jan, Niu, Xing, Ojha, Atul Kr., Ortega, John, Papi, Sara, Polák, Peter, Pospíšil, Adam, Pecina, Pavel, Salesky, Elizabeth, Sethiya, Nivedita, Sarkar, Balaram, Shi, Jiatong, Sikasote, Claytone, Sperber, Matthias, Stüker, Sebastian, Sudoh, Katsuhito, Thompson, Brian, Turchi, Marco, Waibel, Alex, Watanabe, Shinji, Wilken, Patrick, Zemánek, Petr, Zevallos, Rodolfo
This paper reports on the shared tasks organized by the 21st IWSLT Conference. The shared tasks address 7 scientific challenges in spoken language translation: simultaneous and offline translation, automatic subtitling and dubbing, speech-to-speech t
Externí odkaz:
http://arxiv.org/abs/2411.05088
Speech disfluency commonly occurs in conversational and spontaneous speech. However, standard Automatic Speech Recognition (ASR) models struggle to accurately recognize these disfluencies because they are typically trained on fluent transcripts. Curr
Externí odkaz:
http://arxiv.org/abs/2409.10177
Direct speech translation (ST) models often struggle with rare words. Incorrect translation of these words can have severe consequences, impacting translation quality and user trust. While rare word translation is inherently challenging for neural mo
Externí odkaz:
http://arxiv.org/abs/2409.09009
Bayesian observer and actor models have provided normative explanations for many behavioral phenomena in perception, sensorimotor control, and other areas of cognitive science and neuroscience. They attribute behavioral variability and biases to diff
Externí odkaz:
http://arxiv.org/abs/2409.03710
Plug, Play, and Fuse: Zero-Shot Joint Decoding via Word-Level Re-ranking Across Diverse Vocabularies
Recent advancements in NLP have resulted in models with specialized strengths, such as processing multimodal inputs or excelling in specific domains. However, real-world tasks, like multimodal translation, often require a combination of these strengt
Externí odkaz:
http://arxiv.org/abs/2408.11327
Autor:
Koneru, Sai, Nguyen, Thai-Binh, Pham, Ngoc-Quan, Liu, Danni, Li, Zhaolin, Waibel, Alexander, Niehues, Jan
Large Language Models (LLMs) are currently under exploration for various tasks, including Automatic Speech Recognition (ASR), Machine Translation (MT), and even End-to-End Speech Translation (ST). In this paper, we present KIT's offline submission in
Externí odkaz:
http://arxiv.org/abs/2406.16777
Autor:
Dinh, Tu Anh, Mullov, Carlos, Bärmann, Leonard, Li, Zhaolin, Liu, Danni, Reiß, Simon, Lee, Jueun, Lerzer, Nathan, Ternava, Fabian, Gao, Jianfeng, Röddiger, Tobias, Waibel, Alexander, Asfour, Tamim, Beigl, Michael, Stiefelhagen, Rainer, Dachsbacher, Carsten, Böhm, Klemens, Niehues, Jan
With the rapid development of Large Language Models (LLMs), it is crucial to have benchmarks which can evaluate the ability of LLMs on different domains. One common use of LLMs is performing tasks on scientific topics, such as writing algorithms, que
Externí odkaz:
http://arxiv.org/abs/2406.10421