Search Results for author:
Found 50 papers, 25 papers with code
Improving Whisper's Recognition Performance for Under-Represented Language Kazakh Leveraging Unpaired Speech and Text
It is worth researching how to utilize low-cost data to improve the performance of Whisper on under-represented languages.
Human-like object concept representations emerge naturally in multimodal large language models
The conceptualization and categorization of natural objects in the human mind have long intrigued cognitive scientists and neuroscientists, offering crucial insights into human perception and cognition.
GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinement
Notably, ASR models trained on GigaSpeech 2 can reduce the word error rate for Thai, Indonesian, and Vietnamese on our challenging and realistic YouTube test set by 25% to 40% compared to the Whisper large-v3 model, with merely 10% model parameters.
E-Bench: Towards Evaluating the Ease-of-Use of Large Language Models
Most large language models (LLMs) are sensitive to prompts, and another synonymous expression or a typo may lead to unexpected results for the model.
Towards Detecting LLMs Hallucination via Markov Chain-based Multi-agent Debate Framework
The advent of large language models (LLMs) has facilitated the development of natural language text generation.
SFANet: Spatial-Frequency Attention Network for Weather Forecasting
Weather forecasting plays a critical role in various sectors, driving decision-making and risk management.
Parallel Decoding via Hidden Transfer for Lossless Large Language Model Acceleration
In this paper, we propose a novel parallel decoding approach, namely \textit{hidden transfer}, which decodes multiple successive tokens simultaneously in a single forward pass.
StyleChat: Learning Recitation-Augmented Memory in LLMs for Stylized Dialogue Generation
Furthermore, although many prompt-based methods have been proposed to accomplish specific tasks, their performance in complex real-world scenarios involving a wide variety of dialog styles further enhancement.
SpeechColab Leaderboard: An Open-Source Platform for Automatic Speech Recognition Evaluation
In this paper we introduce the SpeechColab Leaderboard, a general-purpose, open-source platform designed for ASR evaluation.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
Progressive Conservative Adaptation for Evolving Target Domains
Moreover, as adjusting to the most recent target domain can interfere with the features learned from previous target domains, we develop a conservative sparse attention mechanism.
SignVTCL: Multi-Modal Continuous Sign Language Recognition Enhanced by Visual-Textual Contrastive Learning
Sign language recognition (SLR) plays a vital role in facilitating communication for the hearing-impaired community.
Leveraging Frequency Domain Learning in 3D Vessel Segmentation
In this study, we leverage Fourier domain learning as a substitute for multi-scale convolutional kernels in 3D hierarchical segmentation models, which can reduce computational expenses while preserving global receptive fields within the network.
Distribution-Aware Calibration for Object Detection with Noisy Bounding Boxes
Unfortunately, the resultant noisy bounding boxes could cause corrupt supervision signals and thus diminish detection performance.
CharacterChat: Learning towards Conversational AI with Personalized Social Support
In our modern, fast-paced, and interconnected world, the importance of mental well-being has grown into a matter of great urgency.
DialoGPS: Dialogue Path Sampling in Continuous Semantic Space for Data Augmentation in Multi-Turn Conversations
In open-domain dialogue generation tasks, contexts and responses in most datasets are one-to-one mapped, violating an important many-to-many characteristic: a context leads to various responses, and a response answers multiple contexts.
3DSAM-adapter: Holistic Adaptation of SAM from 2D to 3D for Promptable Medical Image Segmentation
Notably, the original SAM architecture is designed for 2D natural images, therefore would not be able to extract the 3D spatial information from volumetric medical data effectively.
VSTAR: A Video-grounded Dialogue Dataset for Situated Semantic Understanding with Scene and Topic Transitions
Video-grounded dialogue understanding is a challenging problem that requires machine to perceive, parse and reason over situated semantics extracted from weakly aligned video and dialogues.
Cross-lingual Alzheimer's Disease detection based on paralinguistic and pre-trained features
We present our submission to the ICASSP-SPGC-2023 ADReSS-M Challenge Task, which aims to investigate which acoustic features can be generalized and transferred across languages for Alzheimer's Disease (AD) prediction.
PointPatchMix: Point Cloud Mixing with Patch Scoring
With Point-MAE as our baseline, our model surpasses previous methods by a significant margin, achieving 86. 3% accuracy on ScanObjectNN and 94. 1% accuracy on ModelNet40.
Recognizing Nested Entities from Flat Supervision: A New NER Subtask, Feasibility and Challenges
However, existing nested NER models heavily rely on training data annotated with nested entities, while labeling such data is costly.
Decoding Visual Neural Representations by Multimodal Learning of Brain-Visual-Linguistic Features
Finally, we construct three trimodal matching datasets, and the extensive experiments lead to some interesting conclusions and cognitive insights: 1) decoding novel visual categories from human brain activity is practically possible with good accuracy; 2) decoding models using the combination of visual and linguistic features perform much better than those using either of them alone; 3) visual perception may be accompanied by linguistic influences to represent the semantics of visual stimuli.
Deep Span Representations for Named Entity Recognition
However, this typically results in significant ineffectiveness for long-span entities, a coupling between the representations of overlapping spans, and ultimately a performance degradation.
Structure Regularized Attentive Network for Automatic Femoral Head Necrosis Diagnosis and Localization
However, due to the tissue overlap, X-ray images are difficult to provide fine-grained features for early diagnosis.
Spiral Contrastive Learning: An Efficient 3D Representation Learning Method for Unannotated CT Lesions
In the past five years, several works have tailored for unsupervised representations of CT lesions via two-dimensional (2D) and three-dimensional (3D) self-supervised learning (SSL) algorithms.
CM-MLP: Cascade Multi-scale MLP with Axial Context Relation Encoder for Edge Segmentation of Medical Image
However, those methods have the following challenges when dealing with the edges of the medical images: (1) Previous convolutional-based methods do not focus on the boundary relationship between foreground and background around the segmentation edge, which leads to the degradation of segmentation performance when the edge changes complexly.
Faint Features Tell: Automatic Vertebrae Fracture Screening Assisted by Contrastive Learning
Long-term vertebral fractures severely affect the life quality of patients, causing kyphotic, lumbar deformity and even paralysis.
Deep 3D Vessel Segmentation based on Cross Transformer Network
In CTN, a transformer module is constructed in parallel to a U-Net to learn long-distance dependencies between different anatomical regions; and these dependencies are communicated to the U-Net at multiple stages to endow it with global awareness.
RePFormer: Refinement Pyramid Transformer for Robust Facial Landmark Detection
This paper presents a Refinement Pyramid Transformer (RePFormer) for robust facial landmark detection.
Computer-aided Tuberculosis Diagnosis with Attribute Reasoning Assistance
Although deep learning algorithms have been intensively developed for computer-aided tuberculosis diagnosis (CTD), they mainly depend on carefully annotated datasets, leading to much time and resource consumption.
The THUEE System Description for the IARPA OpenASR21 Challenge
For the Constrained training condition, we construct our basic ASR system based on the standard hybrid architecture.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+4
Boundary Smoothing for Named Entity Recognition
Neural named entity recognition (NER) models may easily encounter the over-confidence issue, which degrades the performance and calibration.
Ranked #2 on
Chinese Named Entity Recognition
on Weibo NER
Chinese Named Entity Recognition
named-entity-recognition
+3
Benchmarking Domain Generalization on EEG-based Emotion Recognition
Specifically, numerous domain adaptation (DA) algorithms have been exploited in the past five years to enhance the generalization of emotion recognition models across subjects.
A Unified Framework of Medical Information Annotation and Extraction for Chinese Clinical Text
The resulted annotated corpus includes 1, 200 full medical records (or 18, 039 broken-down documents), and achieves inter-annotator agreements (IAAs) of 94. 53%, 73. 73% and 91. 98% F 1 scores for the three tasks.
Pedestrian Detection: Domain Generalization, CNNs, Transformers and Beyond
As for the data, we show that the autonomous driving benchmarks are monotonous in nature, that is, they are not diverse in scenarios and dense in pedestrians.
Stylized Dialogue Generation with Multi-Pass Dual Learning
Considering there is no parallel data between the contexts and the responses of target style S1, existing works mainly use back translation to generate stylized synthetic data for training, where the data about context, target style S1 and an intermediate style S0 is used.
Efficient Person Search: An Anchor-Free Approach
Third, by investigating the advantages of both anchor-based and anchor-free models, we further augment AlignPS with an ROI-Align head, which significantly improves the robustness of re-id features while still keeping our model highly efficient.
Ranked #4 on
Person Search
on PRW
MVCNet: Multiview Contrastive Network for Unsupervised Representation Learning for 3D CT Lesions
To avoid the subjectivity associated with these methods, we propose the MVCNet, a novel unsupervised three dimensional (3D) representation learning method working in a transformation-free manner.
Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers
Unlike existing CNN-based methods, we adopt a transformer encoder, which learns more powerful and robust representations.
Ranked #10 on
Medical Image Segmentation
on CVC-ColonDB
MS-MDA: Multisource Marginal Distribution Adaptation for Cross-subject and Cross-session EEG Emotion Recognition
Although several studies have adopted domain adaptation (DA) approaches to tackle this problem, most of them treat multiple EEG data from different subjects and sessions together as a single source domain for transfer, which either fails to satisfy the assumption of domain adaptation that the source has a certain marginal distribution, or increases the difficulty of adaptation.
GREN: Graph-Regularized Embedding Network for Weakly-Supervised Disease Localization in X-ray Images
To model the relationship, we propose the Graph Regularized Embedding Network (GREN), which leverages the intra-image and inter-image information to locate diseases on chest X-ray images.
Local-to-Global Self-Attention in Vision Transformers
Transformers have demonstrated great potential in computer vision tasks.
Exploring Visual Context for Weakly Supervised Person Search
This paper inventively considers weakly supervised person search with only bounding box annotations.
When Liebig's Barrel Meets Facial Landmark Detection: A Practical Model
To this end, we first propose a baseline model equipped with one transformer decoder as detection head.
Ranked #5 on
Face Alignment
on COFW
Anchor-Free Person Search
Person search aims to simultaneously localize and identify a query person from realistic, uncropped images, which can be regarded as the unified task of pedestrian detection and person re-identification (re-id).
Ranked #10 on
Person Search
on CUHK-SYSU
Investigating Critical Risk Factors in Liver Cancer Prediction
We exploit liver cancer prediction model using machine learning algorithms based on epidemiological data of over 55 thousand peoples from 2014 to the present.
FOIT: Fast Online Instance Transfer for Improved EEG Emotion Recognition
However, most of them are iterative methods, which need considerable training time and are unfeasible in practice.
Cross Chest Graph for Disease Diagnosis with Structural Relational Reasoning
Locating lesions is important in the computer-aided diagnosis of X-ray images.
Generalizable Pedestrian Detection: The Elephant In The Room
Furthermore, we illustrate that diverse and dense datasets, collected by crawling the web, serve to be an efficient source of pre-training for pedestrian detection.
Ranked #3 on
Pedestrian Detection
on CityPersons
(using extra training data)
DuBox: No-Prior Box Objection Detection via Residual Dual Scale Detectors
Working with multi-scale features, the designed dual scale residual unit makes dual scale detectors no longer run independently.
Semi-supervised Bayesian Deep Multi-modal Emotion Recognition
In this paper, we first build a multi-view deep generative model to simulate the generative process of multi-modality emotional data.
