Related papers: Soundness does not come for free (if at all)
The ultimate goal of verification is to guarantee the safety of deployed neural networks. Here, we claim that all the state-of-the-art verifiers we are aware of fail to reach this goal. Our key insight is that theoretical soundness…
Speech is promising as an objective, convenient tool to monitor health remotely over time using mobile devices. Numerous paralinguistic features have been demonstrated to contain salient information related to an individual's health.…
This paper reports on the design and outcomes of the ICASSP SP Clarity Challenge: Speech Enhancement for Hearing Aids. The scenario was a listener attending to a target speaker in a noisy, domestic environment. There were multiple…
Earable devices, wearables positioned in or around the ear, are undergoing a rapid transformation from audio-centric accessories into multifunctional systems for interaction, contextual awareness, and health monitoring. This evolution is…
This paper discusses {\ae}sthetic issues of sonifications and the relationships between sonification (ars informatica) and music & sound art (ars musica). It is posited that many sonifications have suffered from poor internal ecological…
Audio-text relevance learning refers to learning the shared semantic properties of audio samples and textual descriptions. The standard approach uses binary relevances derived from pairs of audio samples and their human-provided captions,…
While efficient architectures and a plethora of augmentations for end-to-end image classification tasks have been suggested and heavily investigated, state-of-the-art techniques for audio classifications still rely on numerous…
We reply to the comments on our previous paper J. Phys. Soc. Jpn. 91, 024001 (2022)
Neural network (NN) verification aims to formally verify properties of NNs, which is crucial for ensuring the behavior of NN-based models in safety-critical applications. In recent years, the community has developed many NN verifiers and…
Are the sciences not advancing at an ever increasing speed? We contrast this popular perspective with the view that science funding may actually see diminishing returns, at least regarding established fields. In order to stimulate a larger…
This paper is a continuation of our earlier work "[T. Jin, Y.Y. Li and J. Xiong, On a fractional Nirenberg problem, part I: blow up analysis and compactness of solutions, to appear in J. Eur. Math. Soc.]", where compactness results were…
Recent advances in spoken language understanding benefited from Self-Supervised models trained on large speech corpora. For French, the LeBenchmark project has made such models available and has led to impressive progress on several tasks…
Compositionality has traditionally been understood as a major factor in productivity of language and, more broadly, human cognition. Yet, recently, some research started to question its status, showing that artificial neural networks are…
Improvements in large language models have led to increasing optimism that they can serve as reliable evaluators of natural language generation outputs. In this paper, we challenge this optimism by thoroughly re-evaluating five…
Some misleading statements in a recent paper in PRE are corrected.
In this note we briefly comment a paper by Itin, Obukhov and Hehl criticising our previous paper. We show that all remarks by our critics are ill conceived or irrelevant to our approach and moreover we provide some pertinent new comments to…
Several variations of the Heisenberg uncertainty inequality are derived on the basis of "noise-resolution duality" recently proposed by the authors. The same approach leads to a related inequality that provides an upper limit for the…
In a series of comments, Bonder et al. criticized our work on decoherence due to time dilation [Nature Physics 11, 668-672 (2015)]. First the authors erroneously claimed that our results contradict the equivalence principle, only to…
A by-no-means-complete collection of references for those interested in intonational meaning, with other miscellaneous references on intonation included. Additional references are welcome, and should be sent to [email protected].
There is a growing abundance of publicly available or company-owned audio/video archives, highlighting the increasing importance of efficient access to desired content and information retrieval from these archives. This paper investigates…