Related papers: Amortized Context Vector Inference for Sequence-to…
The spatial attention mechanism captures long-range dependencies by aggregating global contextual information to each query location, which is beneficial for semantic segmentation. In this paper, we present a sparse spatial attention…
Developments in machine learning interpretability techniques over the past decade have provided new tools to observe the image regions that are most informative for classification and localization in artificial neural networks (ANNs). Are…
Recent decoder-only autoregressive text-to-speech (AR-TTS) models produce high-fidelity speech, but their memory and compute costs scale quadratically with sequence length due to full self-attention. In this paper, we propose WAND, Windowed…
Transformers excel at sequence modeling but face quadratic complexity, while linear attention offers improved efficiency but often compromises recall accuracy over long contexts. In this work, we introduce Native Hybrid Attention (NHA), a…
This paper proposes a non-autoregressive extension of our previously proposed sequence-to-sequence (S2S) model-based voice conversion (VC) methods. S2S model-based VC methods have attracted particular attention in recent years for their…
An end-to-end (E2E) ASR model implicitly learns a prior Internal Language Model (ILM) from the training transcripts. To fuse an external LM using Bayes posterior theory, the log likelihood produced by the ILM has to be accurately estimated…
Automatic Speech Recognition (ASR) still face challenges when recognizing time-variant rare-phrases. Contextual biasing (CB) modules bias ASR model towards such contextually-relevant phrases. During training, a list of biasing phrases are…
The internal workings of modern deep learning models stay often unclear to an external observer, although spatial attention mechanisms are involved. The idea of this work is to translate these spatial attentions into natural language to…
We propose the notion of Attention-Aware Visualizations (AAVs) that track the user's perception of a visual representation over time and feed this information back to the visualization. Such context awareness is particularly useful for…
Current attention algorithms (e.g., self-attention) are stimulus-driven and highlight all the salient objects in an image. However, intelligent agents like humans often guide their attention based on the high-level task at hand, focusing…
Context-aware Machine Translation aims to improve translations of sentences by incorporating surrounding sentences as context. Towards this task, two main architectures have been applied, namely single-encoder (based on concatenation) and…
Neural contextual biasing allows speech recognition models to leverage contextually relevant information, leading to improved transcription accuracy. However, the biasing mechanism is typically based on a cross-attention module between the…
The Transformer self-attention network has recently shown promising performance as an alternative to recurrent neural networks (RNNs) in end-to-end (E2E) automatic speech recognition (ASR) systems. However, the Transformer has a drawback in…
Neural sequence-to-sequence models are currently the dominant approach in several natural language processing tasks, but require large parallel corpora. We present a sequence-to-sequence-to-sequence autoencoder (SEQ^3), consisting of two…
Graphs arise across diverse domains, from biology and chemistry to social and information networks, as well as in transportation and logistics. Inference on graph-structured data requires methods that are permutation-invariant, scalable…
Attention-based sequence-to-sequence models have shown promising results in automatic speech recognition. Using these architectures, one-dimensional input and output sequences are related by an attention approach, thereby replacing more…
This paper aims to address few-shot segmentation. While existing prototype-based methods have achieved considerable success, they suffer from uncertainty and ambiguity caused by limited labeled examples. In this work, we propose attentional…
The quadratic complexity of attention remains the central bottleneck in long-context inference for large language models. Prior acceleration methods either sparsify the attention map with structured patterns or permanently evict tokens at…
By incorporating additional contextual information, deep biasing methods have emerged as a promising solution for speech recognition of personalized words. However, for real-world voice assistants, always biasing on such personalized words…
Transformer models are revolutionizing machine learning, but their inner workings remain mysterious. In this work, we present a new visualization technique designed to help researchers understand the self-attention mechanism in transformers…