Error correction techniques have been used to refine the output sentences from automatic speech recognition (ASR) models and achieve a lower word error rate (WER) than original ASR outputs. Previous works usually use a sequence-to-sequence model to correct an ASR output sentence autoregressively, which causes large latency and cannot be deployed in online ASR services. A straightforward solution to reduce latency, inspired by non-autoregressive (NAR) neural machine translation, is to use an NAR sequence generation model for ASR error correction, which, however, comes at the cost of significantly increased ASR error rate. In this paper, observing distinctive error patterns and correction operations (i.e., insertion, deletion, and substitution) in ASR, we propose FastCorrect, a novel NAR error correction model based on edit alignment. In training, FastCorrect aligns each source token from an ASR output sentence to the target tokens from the corresponding ground-truth sentence based on the edit distance between the source and target sentences, and extracts the number of target tokens corresponding to each source token during edition/correction, which is then used to train a length predictor and to adjust the source tokens to match the length of the target sentence for parallel generation. In inference, the token number predicted by the length predictor is used to adjust the source tokens for target sequence generation. Experiments on the public AISHELL-1 dataset and an internal industrial-scale ASR dataset show the effectiveness of FastCorrect for ASR error correction: 1) it speeds up the inference by 6-9 times and maintains the accuracy (8-14% WER reduction) compared with the autoregressive correction model; and 2) it outperforms the popular NAR models adopted in neural machine translation and text edition by a large margin.

This work was conducted at Microsoft.

Automated pose correction remains a significant challenge in AI-driven fitness systems, despite extensive research in activity recognition. This work presents PosePilot, a novel system that integrates pose recognition with real-time personalized corrective feedback, overcoming the limitations of traditional fitness solutions. Using Yoga, a discipline requiring precise spatio-temporal alignment as a case study, we demonstrate PosePilot's ability to analyze complex physical movements. Designed for deployment on edge devices, PosePilot can be extended to various at-home and outdoor exercises. We employ a Vanilla LSTM, allowing the system to capture temporal dependencies for pose recognition. Additionally, a BiLSTM with multi-head Attention enhances the model's ability to process motion contexts, selectively focusing on key limb angles for accurate error detection while maintaining computational efficiency. As part of this work, we introduce a high-quality video dataset used for evaluating our models. Most importantly, PosePilot provides instant corrective feedback at every stage of a movement, ensuring precise posture adjustments throughout the exercise routine. The proposed approach 1) performs automatic human posture recognition, 2) provides personalized posture correction feedback at each instant which is crucial in Yoga, and 3) offers a lightweight and robust posture correction model feasible for deploying on edge devices in real-world environments.

Open set label shift (OSLS) occurs when label distributions change from a source to a target distribution, and the target distribution has an additional out-of-distribution (OOD) class. In this work, we build estimators for both source and target open set label distributions using a source domain in-distribution (ID) classifier and an ID/OOD classifier . With reasonable assumptions on the ID/OOD classifier, the estimators are assembled into a sequence of three stages: 1) an estimate of the source label distribution of the OOD class, 2) an EM algorithm for Maximum Likelihood estimates (MLE) of the target label distribution, and 3) an estimate of the target label distribution of OOD class under relaxed assumptions on the OOD classifier . The sampling errors of estimates in 1) and 3) are quantified with a concentration inequality. The estimation result allows us to correct the ID classifier trained on the source distribution to the target distribution without retraining. Experiments on a variety of open set label shift settings demonstrate the effectiveness of our model.

Numerical weather prediction (NWP) is crucial in weather forecasting, providing indispensable guidance across temporal scales from nowcasting to seasonal forecasting (Bauer et al., 2015). As society becomes more dependent on accurate forecasts, there is an increasing demand for high-quality predictions, particularly in extreme events such as heat waves and cold surges, which can have severe social and economic impacts(Br as et al., 2023; Miao et al., 2024). Furthermore, atmospheric forecasts serve as critical boundary conditions for coupled Earth system models, where their accuracy directly governs the predictive capabilities of oceanographic and cryospheric simulations through dynamic coupling mechanisms. While the ECMWF's Integrated Forecasting System (IFS) represents the state-of-the-art in global operational prediction (Molteni et al., 1996), persistent systematic biases still exist, which arise from three fundamental sources: (1) inadequate spatial resolution to resolve subgrid-scale processes (Mishra et al., 2021), (2) inherent limitations in physical parameterization schemes (Berner et al., 2017; Brenowitz & Bretherton, 2018), and (3) uncertainties in initial/boundary condition specification (Peng & Xie, 2006). Current bias correction paradigms predominantly employ statistical postprocessing techniques, including uni-variate regression frameworks (Turco et al., 2017), adaptive filtering techniques (Chandramouli et al., 2022), and probabilistic calibration methods (Yumnam et al., 2022).

The Chinese Spelling Correction (CSC) task focuses on detecting and correcting spelling errors in sentences. Current research primarily explores two approaches: traditional multimodal pre-trained models and large language models (LLMs). However, LLMs face limitations in CSC, particularly over-correction, making them suboptimal for this task. While existing studies have investigated the use of phonetic and graphemic information in multimodal CSC models, effectively leveraging these features to enhance correction performance remains a challenge. To address this, we propose the Multimodal Analysis for Character Usage (\textbf{MACU}) experiment, identifying potential improvements for multimodal correctison. Based on empirical findings, we introduce \textbf{NamBert}, a novel multimodal model for Chinese spelling correction. Experiments on benchmark datasets demonstrate NamBert's superiority over SOTA methods. We also conduct a comprehensive comparison between NamBert and LLMs, systematically evaluating their strengths and limitations in CSC. Our code and model are available at https://github.com/iioSnail/NamBert.

Embedding inversion, i.e., reconstructing text given its embedding and black-box access to the embedding encoder, is a fundamental problem in both NLP and security. From the NLP perspective, it helps determine how much semantic information about the input is retained in the embedding. From the security perspective, it measures how much information is leaked by vector databases and embedding-based retrieval systems. State-of-the-art methods for embedding inversion, such as vec2text, have high accuracy but require (a) training a separate model for each embedding, and (b) a large number of queries to the corresponding encoder. We design, implement, and evaluate ZSInvert, a zero-shot inversion method based on the recently proposed adversarial decoding technique. ZSInvert is fast, query-efficient, and can be used for any text embedding without training an embedding-specific inversion model. We measure the effectiveness of ZSInvert on several embeddings and demonstrate that it recovers key semantic information about the corresponding texts.

PET is a functional imaging method that visualizes metabolic processes. TOF information can be derived from coincident detector signals and incorporated into image reconstruction to enhance the SNR. PET detectors are typically assessed by their CTR, but timing performance is degraded by various factors. Research on timing calibration seeks to mitigate these degradations and restore accurate timing information. While many calibration methods use analytical approaches, machine learning techniques have recently gained attention due to their flexibility. We developed a residual physics-based calibration approach that combines prior domain knowledge with the power of machine learning models. This approach begins with an initial analytical calibration addressing first-order skews. The remaining deviations, regarded as residual effects, are used to train machine learning models to eliminate higher-order skews. The key advantage is that the experimenter guides the learning process through the definition of timing residuals. In earlier studies, we developed models that directly predicted the expected time difference, which offered corrections only implicitly (implicit correction models). In this study, we introduce a new definition for timing residuals, enabling us to train models that directly predict correction values (explicit correction models). The explicit correction approach significantly simplifies data acquisition, improves linearity, and enhances timing performance from $371 \pm 6$ ps to $281 \pm 5$ ps for coincidences from 430 keV to 590 keV. Additionally, the new definition reduces model size, making it suitable for high-throughput applications like PET scanners. Experiments were conducted using two detector stacks composed of $4 \times 4$ LYSO:Ce,Ca crystals ($3.8\times 3.8\times 20$ mm$^{3}$) coupled to $4 \times 4$ Broadcom NUV-MT SiPMs and digitized with the TOFPET2 ASIC.

Synthetic data generation is widely recognized as a way to enhance the quality of neural grammatical error correction (GEC) systems. However, current approaches often lack diversity or are too simplistic to generate the wide range of grammatical errors made by humans, especially for low-resource languages such as Arabic. In this paper, we will develop the error tagging model and the synthetic data generation model to create a large synthetic dataset in Arabic for grammatical error correction. In the error tagging model, the correct sentence is categorized into multiple error types by using the DeBERTav3 model. Arabic Error Type Annotation tool (ARETA) is used to guide multi-label classification tasks in an error tagging model in which each sentence is classified into 26 error tags. The synthetic data generation model is a back-translation-based model that generates incorrect sentences by appending error tags before the correct sentence that was generated from the error tagging model using the ARAT5 model. In the QALB-14 and QALB-15 Test sets, the error tagging model achieved 94.42% F1, which is state-of-the-art in identifying error tags in clean sentences. As a result of our syntactic data training in grammatical error correction, we achieved a new state-of-the-art result of F1-Score: 79.36% in the QALB-14 Test set. We generate 30,219,310 synthetic sentence pairs by using a synthetic data generation model.