Qing He

Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri et al. · allen-ai, berkeley

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Yiqing Wu, Ruobing Xie, Yongchun Zhu et al.

Multi-behavior recommendation (MBR) aims to jointly consider multiple behaviors to improve the target behavior's performance. We argue that MBR models should: (1) model the coarse-grained commonalities between different behaviors of a user, (2) consider both individual sequence view and global graph view in multi-behavior modeling, and (3) capture the fine-grained differences between multiple behaviors of a user. In this work, we propose a novel Multi-behavior Multi-view Contrastive Learning Recommendation (MMCLR) framework, including three new CL tasks to solve the above challenges, respectively. The multi-behavior CL aims to make different user single-behavior representations of the same user in each view to be similar. The multi-view CL attempts to bridge the gap between a user's sequence-view and graph-view representations. The behavior distinction CL focuses on modeling fine-grained differences of different behaviors. In experiments, we conduct extensive evaluations and ablation tests to verify the effectiveness of MMCLR and various CL tasks on two real-world datasets, achieving SOTA performance over existing baselines. Our code will be available on \url{https://github.com/wyqing20/MMCLR}

Kuan Li, Yang Liu, Xiang Ao et al.

Benefiting from the message passing mechanism, Graph Neural Networks (GNNs) have been successful on flourish tasks over graph data. However, recent studies have shown that attackers can catastrophically degrade the performance of GNNs by maliciously modifying the graph structure. A straightforward solution to remedy this issue is to model the edge weights by learning a metric function between pairwise representations of two end nodes, which attempts to assign low weights to adversarial edges. The existing methods use either raw features or representations learned by supervised GNNs to model the edge weights. However, both strategies are faced with some immediate problems: raw features cannot represent various properties of nodes (e.g., structure information), and representations learned by supervised GNN may suffer from the poor performance of the classifier on the poisoned graph. We need representations that carry both feature information and as mush correct structure information as possible and are insensitive to structural perturbations. To this end, we propose an unsupervised pipeline, named STABLE, to optimize the graph structure. Finally, we input the well-refined graph into a downstream classifier. For this part, we design an advanced GCN that significantly enhances the robustness of vanilla GCN without increasing the time complexity. Extensive experiments on four real-world graph benchmarks demonstrate that STABLE outperforms the state-of-the-art methods and successfully defends against various attacks.

Shuokai Li, Ruobing Xie, Yongchun Zhu et al.

Conversational recommender systems (CRS) aim to provide highquality recommendations in conversations. However, most conventional CRS models mainly focus on the dialogue understanding of the current session, ignoring other rich multi-aspect information of the central subjects (i.e., users) in recommendation. In this work, we highlight that the user's historical dialogue sessions and look-alike users are essential sources of user preferences besides the current dialogue session in CRS. To systematically model the multi-aspect information, we propose a User-Centric Conversational Recommendation (UCCR) model, which returns to the essence of user preference learning in CRS tasks. Specifically, we propose a historical session learner to capture users' multi-view preferences from knowledge, semantic, and consuming views as supplements to the current preference signals. A multi-view preference mapper is conducted to learn the intrinsic correlations among different views in current and historical sessions via self-supervised objectives. We also design a temporal look-alike user selector to understand users via their similar users. The learned multi-aspect multi-view user preferences are then used for the recommendation and dialogue generation. In experiments, we conduct comprehensive evaluations on both Chinese and English CRS datasets. The significant improvements over competitive models in both recommendation and dialogue generation verify the superiority of UCCR.

Shuokai Li, Yongchun Zhu, Ruobing Xie et al. · utoronto

Conversational recommender systems (CRS) aim to capture user's current intentions and provide recommendations through real-time multi-turn conversational interactions. As a human-machine interactive system, it is essential for CRS to improve the user experience. However, most CRS methods neglect the importance of user experience. In this paper, we propose two key points for CRS to improve the user experience: (1) Speaking like a human, human can speak with different styles according to the current dialogue context. (2) Identifying fine-grained intentions, even for the same utterance, different users have diverse finegrained intentions, which are related to users' inherent preference. Based on the observations, we propose a novel CRS model, coined Customized Conversational Recommender System (CCRS), which customizes CRS model for users from three perspectives. For human-like dialogue services, we propose multi-style dialogue response generator which selects context-aware speaking style for utterance generation. To provide personalized recommendations, we extract user's current fine-grained intentions from dialogue context with the guidance of user's inherent preferences. Finally, to customize the model parameters for each user, we train the model from the meta-learning perspective. Extensive experiments and a series of analyses have shown the superiority of our CCRS on both the recommendation and dialogue services.

Mumin Jin, Prashant Serai, Jilong Wu et al.

Most people who have tried to learn a foreign language would have experienced difficulties understanding or speaking with a native speaker's accent. For native speakers, understanding or speaking a new accent is likewise a difficult task. An accent conversion system that changes a speaker's accent but preserves that speaker's voice identity, such as timbre and pitch, has the potential for a range of applications, such as communication, language learning, and entertainment. Existing accent conversion models tend to change the speaker identity and accent at the same time. Here, we use adversarial learning to disentangle accent dependent features while retaining other acoustic characteristics. What sets our work apart from existing accent conversion models is the capability to convert an unseen speaker's utterance to multiple accents while preserving its original voice identity. Subjective evaluations show that our model generates audio that sound closer to the target accent and like the original speaker.

Wenhao Ding, Qing He, Hanghang Tong et al.

As a fundamental mathmatical tool in many engineering disciplines, coupled differential equation groups are being widely used to model complex structures containing multiple physical quantities. Engineers constantly adjust structural parameters at the design stage, which requires a highly efficient solver. The rise of deep learning technologies has offered new perspectives on this task. Unfortunately, existing black-box models suffer from poor accuracy and robustness, while the advanced methodologies of single-output operator regression cannot deal with multiple quantities simultaneously. To address these challenges, we propose PINO-CDE, a deep learning framework for solving coupled differential equation groups (CDEs) along with an equation normalization algorithm for performance enhancing. Based on the theory of physics-informed neural operator (PINO), PINO-CDE uses a single network for all quantities in a CDEs, instead of training dozens, or even hundreds of networks as in the existing literature. We demonstrate the flexibility and feasibility of PINO-CDE for one toy example and two engineering applications: vehicle-track coupled dynamics (VTCD) and reliability assessment for a four-storey building (uncertainty propagation). The performance of VTCD indicates that PINO-CDE outperforms existing software and deep learning-based methods in terms of efficiency and precision, respectively. For the uncertainty propagation task, PINO-CDE provides higher-resolution results in less than a quarter of the time incurred when using the probability density evolution method (PDEM). This framework integrates engineering dynamics and deep learning technologies and may reveal a new concept for CDEs solving and uncertainty propagation.

Mingyang Liu, Li Xiao, Huiqin Jiang et al.

The spread of the novel coronavirus disease 2019 (COVID-19) has claimed millions of lives. Automatic segmentation of lesions from CT images can assist doctors with screening, treatment, and monitoring. However, accurate segmentation of lesions from CT images can be very challenging due to data and model limitations. Recently, Transformer-based networks have attracted a lot of attention in the area of computer vision, as Transformer outperforms CNN at a bunch of tasks. In this work, we propose a novel network structure that combines CNN and Transformer for the segmentation of COVID-19 lesions. We further propose an efficient semi-supervised learning framework to address the shortage of labeled data. Extensive experiments showed that our proposed network outperforms most existing networks and the semi-supervised learning framework can outperform the base network by 3.0% and 8.2% in terms of Dice coefficient and sensitivity.

Philipp Klumpp, Pooja Chitkara, Leda Sarı et al.

The awareness for biased ASR datasets or models has increased notably in recent years. Even for English, despite a vast amount of available training data, systems perform worse for non-native speakers. In this work, we improve an accent-conversion model (ACM) which transforms native US-English speech into accented pronunciation. We include phonetic knowledge in the ACM training to provide accurate feedback about how well certain pronunciation patterns were recovered in the synthesized waveform. Furthermore, we investigate the feasibility of learned accent representations instead of static embeddings. Generated data was then used to train two state-of-the-art ASR systems. We evaluated our approach on native and non-native English datasets and found that synthetically accented data helped the ASR to better understand speech from seen accents. This observation did not translate to unseen accents, and it was not observed for a model that had been pre-trained exclusively with native speech.

Yancheng Wang, Osama Hanna, Ruiming Xie et al.

Emotion recognition in speech presents a complex multimodal challenge, requiring comprehension of both linguistic content and vocal expressivity, particularly prosodic features such as fundamental frequency, intensity, and temporal dynamics. Although large language models (LLMs) have shown promise in reasoning over textual transcriptions for emotion recognition, they typically neglect fine-grained prosodic information, limiting their effectiveness and interpretability. In this work, we propose VowelPrompt, a linguistically grounded framework that augments LLM-based emotion recognition with interpretable, fine-grained vowel-level prosodic cues. Drawing on phonetic evidence that vowels serve as primary carriers of affective prosody, VowelPrompt extracts pitch-, energy-, and duration-based descriptors from time-aligned vowel segments, and converts these features into natural language descriptions for better interpretability. Such a design enables LLMs to jointly reason over lexical semantics and fine-grained prosodic variation. Moreover, we adopt a two-stage adaptation procedure comprising supervised fine-tuning (SFT) followed by Reinforcement Learning with Verifiable Reward (RLVR), implemented via Group Relative Policy Optimization (GRPO), to enhance reasoning capability, enforce structured output adherence, and improve generalization across domains and speaker variations. Extensive evaluations across diverse benchmark datasets demonstrate that VowelPrompt consistently outperforms state-of-the-art emotion recognition methods under zero-shot, fine-tuned, cross-domain, and cross-linguistic conditions, while enabling the generation of interpretable explanations that are jointly grounded in contextual semantics and fine-grained prosodic structure.

Jason Fong, Yun Wang, Prabhav Agrawal et al.

Text-based voice editing (TBVE) uses synthetic output from text-to-speech (TTS) systems to replace words in an original recording. Recent work has used neural models to produce edited speech that is similar to the original speech in terms of clarity, speaker identity, and prosody. However, one limitation of prior work is the usage of finetuning to optimise performance: this requires further model training on data from the target speaker, which is a costly process that may incorporate potentially sensitive data into server-side models. In contrast, this work focuses on the zero-shot approach which avoids finetuning altogether, and instead uses pretrained speaker verification embeddings together with a jointly trained reference encoder to encode utterance-level information that helps capture aspects such as speaker identity and prosody. Subjective listening tests find that both utterance embeddings and a reference encoder improve the continuity of speaker identity and prosody between the edited synthetic speech and unedited original recording in the zero-shot setting.

Yiran Qiao, Xiang Ao, Jing Chen et al.

The rise of live streaming has transformed online interaction, enabling massive real-time engagement but also exposing platforms to complex risks such as scams and coordinated malicious behaviors. Detecting these risks is challenging because harmful actions often accumulate gradually and recur across seemingly unrelated streams. To address this, we propose CS-VAR (Cross-Session Evidence-Aware Retrieval-Augmented Detector) for live streaming risk assessment. In CS-VAR, a lightweight, domain-specific model performs fast session-level risk inference, guided during training by a Large Language Model (LLM) that reasons over retrieved cross-session behavioral evidence and transfers its local-to-global insights to the small model. This design enables the small model to recognize recurring patterns across streams, perform structured risk assessment, and maintain efficiency for real-time deployment. Extensive offline experiments on large-scale industrial datasets, combined with online validation, demonstrate the state-of-the-art performance of CS-VAR. Furthermore, CS-VAR provides interpretable, localized signals that effectively empower real-world moderation for live streaming.

Yiran Qiao, Jing Chen, Xiang Ao et al.

Live streaming has become a cornerstone of today's internet, enabling massive real-time social interactions. However, it faces severe risks arising from sparse, coordinated malicious behaviors among multiple participants, which are often concealed within normal activities and challenging to detect timely and accurately. In this work, we provide a pioneering study on risk assessment in live streaming rooms, characterized by weak supervision where only room-level labels are available. We formulate the task as a Multiple Instance Learning (MIL) problem, treating each room as a bag and defining structured user-timeslot capsules as instances. These capsules represent subsequences of user actions within specific time windows, encapsulating localized behavioral patterns. Based on this formulation, we propose AC-MIL, an Action-aware Capsule MIL framework that models both individual behaviors and group-level coordination patterns. AC-MIL captures multi-granular semantics and behavioral cues through a serial and parallel architecture that jointly encodes temporal dynamics and cross-user dependencies. These signals are integrated for robust room-level risk prediction, while also offering interpretable evidence at the behavior segment level. Extensive experiments on large-scale industrial datasets from Douyin demonstrate that AC-MIL significantly outperforms MIL and sequential baselines, establishing new state-of-the-art performance in room-level risk assessment for live streaming. Moreover, AC-MIL provides capsule-level interpretability, enabling identification of risky behavior segments as actionable evidence for intervention. The project page is available at: https://qiaoyran.github.io/AC-MIL/.

Yiran Qiao, Xiang Ao, Yang Liu et al.

Recent prevailing works on graph machine learning typically follow a similar methodology that involves designing advanced variants of graph neural networks (GNNs) to maintain the superior performance of GNNs on different graphs. In this paper, we aim to streamline the GNN design process and leverage the advantages of Large Language Models (LLMs) to improve the performance of GNNs on downstream tasks. We formulate a new paradigm, coined "LLMs-as-Consultants," which integrates LLMs with GNNs in an interactive manner. A framework named LOGIN (LLM Consulted GNN training) is instantiated, empowering the interactive utilization of LLMs within the GNN training process. First, we attentively craft concise prompts for spotted nodes, carrying comprehensive semantic and topological information, and serving as input to LLMs. Second, we refine GNNs by devising a complementary coping mechanism that utilizes the responses from LLMs, depending on their correctness. We empirically evaluate the effectiveness of LOGIN on node classification tasks across both homophilic and heterophilic graphs. The results illustrate that even basic GNN architectures, when employed within the proposed LLMs-as-Consultants paradigm, can achieve comparable performance to advanced GNNs with intricate designs. Our codes are available at https://github.com/QiaoYRan/LOGIN.

Yuancheng Wang, Zhenyu Tang, Yun Wang et al.

Speech tokenizers are foundational to speech language models, yet existing approaches face two major challenges: (1) balancing trade-offs between encoding semantics for understanding and acoustics for reconstruction, and (2) achieving low bit rates and low token rates. We propose Speech Diffusion Tokenizer (SiTok), a diffusion autoencoder that jointly learns semantic-rich representations through supervised learning and enables high-fidelity audio reconstruction with diffusion. We scale SiTok to 1.6B parameters and train it on 2 million hours of speech. Experiments show that SiTok outperforms strong baselines on understanding, reconstruction and generation tasks, at an extremely low token rate of $12.5$ Hz and a bit-rate of 200 bits-per-second.

Qing He, Dongsheng Bi, Jianrong Lu et al.

The proliferation of Large Language Models (LLMs) presents transformative potential for healthcare, yet practical deployment is hindered by the absence of frameworks that assess real-world clinical utility. Existing benchmarks test static knowledge, failing to capture the dynamic, application-oriented capabilities required in clinical practice. To bridge this gap, we introduce a Medical LLM Benchmark MLB, a comprehensive benchmark evaluating LLMs on both foundational knowledge and scenario-based reasoning. MLB is structured around five core dimensions: Medical Knowledge (MedKQA), Safety and Ethics (MedSE), Medical Record Understanding (MedRU), Smart Services (SmartServ), and Smart Healthcare (SmartCare). The benchmark integrates 22 datasets (17 newly curated) from diverse Chinese clinical sources, covering 64 clinical specialties. Its design features a rigorous curation pipeline involving 300 licensed physicians. Besides, we provide a scalable evaluation methodology, centered on a specialized judge model trained via Supervised Fine-Tuning (SFT) on expert annotations. Our comprehensive evaluation of 10 leading models reveals a critical translational gap: while the top-ranked model, Kimi-K2-Instruct (77.3% accuracy overall), excels in structured tasks like information extraction (87.8% accuracy in MedRU), performance plummets in patient-facing scenarios (61.3% in SmartServ). Moreover, the exceptional safety score (90.6% in MedSE) of the much smaller Baichuan-M2-32B highlights that targeted training is equally critical. Our specialized judge model, trained via SFT on a 19k expert-annotated medical dataset, achieves 92.1% accuracy, an F1-score of 94.37%, and a Cohen's Kappa of 81.3% for human-AI consistency, validating a reproducible and expert-aligned evaluation protocol. MLB thus provides a rigorous framework to guide the development of clinically viable LLMs.

Qing He, Xiaowei Fu, Lei Zhang

Encrypted traffic classification is a critical task for network security. While deep learning has advanced this field, the occlusion of payload semantics by encryption severely challenges standard modeling approaches. Most existing frameworks rely on static and homogeneous pipelines that apply uniform parameter sharing and static fusion strategies across all inputs. This one-size-fits-all static design is inherently flawed: by forcing structured headers and randomized payloads into a unified processing pipeline, it inevitably entangles the raw protocol signals with stochastic encryption noise, thereby degrading the fine-grained discriminative features. In this paper, we propose TrafficMoE, a framework that breaks through the bottleneck of static modeling by establishing a Disentangle-Filter-Aggregate (DFA) paradigm. Specifically, to resolve the structural between-components conflict, the architecture disentangles headers and payloads using dual-branch sparse Mixture-of-Experts (MoE), enabling modality-specific modeling. To mitigate the impact of stochastic noise, an uncertainty-aware filtering mechanism is introduced to quantify reliability and selectively suppress high-variance representations. Finally, to overcome the limitations of static fusion, a routing-guided strategy aggregates cross-modality features dynamically, that adaptively weighs contributions based on traffic context. With this DFA paradigm, TrafficMoE maximizes representational efficiency by focusing solely on the most discriminative traffic features. Extensive experiments on six datasets demonstrate TrafficMoE consistently outperforms state-of-the-art methods, validating the necessity of heterogeneity-aware modeling in encrypted traffic analysis. The source code is publicly available at https://github.com/Posuly/TrafficMoE_main.

Yongchun Zhu, Fuzhen Zhuang, Jindong Wang et al.

In image classification, it is often expensive and time-consuming to acquire sufficient labels. To solve this problem, domain adaptation often provides an attractive option given a large amount of labeled data from a similar nature but different domain. Existing approaches mainly align the distributions of representations extracted by a single structure and the representations may only contain partial information, e.g., only contain part of the saturation, brightness, and hue information. Along this line, we propose Multi-Representation Adaptation which can dramatically improve the classification accuracy for cross-domain image classification and specially aims to align the distributions of multiple representations extracted by a hybrid structure named Inception Adaptation Module (IAM). Based on this, we present Multi-Representation Adaptation Network (MRAN) to accomplish the cross-domain image classification task via multi-representation alignment which can capture the information from different aspects. In addition, we extend Maximum Mean Discrepancy (MMD) to compute the adaptation loss. Our approach can be easily implemented by extending most feed-forward models with IAM, and the network can be trained efficiently via back-propagation. Experiments conducted on three benchmark image datasets demonstrate the effectiveness of MRAN. The code has been available at https://github.com/easezyc/deep-transfer-learning.

Ehab A. AlBadawy, Andrew Gibiansky, Qing He et al.

Neural vocoders, used for converting the spectral representations of an audio signal to the waveforms, are a commonly used component in speech synthesis pipelines. It focuses on synthesizing waveforms from low-dimensional representation, such as Mel-Spectrograms. In recent years, different approaches have been introduced to develop such vocoders. However, it becomes more challenging to assess these new vocoders and compare their performance to previous ones. To address this problem, we present VocBench, a framework that benchmark the performance of state-of-the art neural vocoders. VocBench uses a systematic study to evaluate different neural vocoders in a shared environment that enables a fair comparison between them. In our experiments, we use the same setup for datasets, training pipeline, and evaluation metrics for all neural vocoders. We perform a subjective and objective evaluation to compare the performance of each vocoder along a different axis. Our results demonstrate that the framework is capable of showing the competitive efficacy and the quality of the synthesized samples for each vocoder. VocBench framework is available at https://github.com/facebookresearch/vocoder-benchmark.

Yongchun Zhu, Zhenwei Tang, Yudan Liu et al.

Cold-start problem is still a very challenging problem in recommender systems. Fortunately, the interactions of the cold-start users in the auxiliary source domain can help cold-start recommendations in the target domain. How to transfer user's preferences from the source domain to the target domain, is the key issue in Cross-domain Recommendation (CDR) which is a promising solution to deal with the cold-start problem. Most existing methods model a common preference bridge to transfer preferences for all users. Intuitively, since preferences vary from user to user, the preference bridges of different users should be different. Along this line, we propose a novel framework named Personalized Transfer of User Preferences for Cross-domain Recommendation (PTUPCDR). Specifically, a meta network fed with users' characteristic embeddings is learned to generate personalized bridge functions to achieve personalized transfer of preferences for each user. To learn the meta network stably, we employ a task-oriented optimization procedure. With the meta-generated personalized bridge function, the user's preference embedding in the source domain can be transformed into the target domain, and the transformed user preference embedding can be utilized as the initial embedding for the cold-start user in the target domain. Using large real-world datasets, we conduct extensive experiments to evaluate the effectiveness of PTUPCDR on both cold-start and warm-start stages. The code has been available at https://github.com/easezyc/WSDM2022-PTUPCDR.

Zijun Sun, Xiaoya Li, Xiaofei Sun et al.

Recent pretraining models in Chinese neglect two important aspects specific to the Chinese language: glyph and pinyin, which carry significant syntax and semantic information for language understanding. In this work, we propose ChineseBERT, which incorporates both the {\it glyph} and {\it pinyin} information of Chinese characters into language model pretraining. The glyph embedding is obtained based on different fonts of a Chinese character, being able to capture character semantics from the visual features, and the pinyin embedding characterizes the pronunciation of Chinese characters, which handles the highly prevalent heteronym phenomenon in Chinese (the same character has different pronunciations with different meanings). Pretrained on large-scale unlabeled Chinese corpus, the proposed ChineseBERT model yields significant performance boost over baseline models with fewer training steps. The porpsoed model achieves new SOTA performances on a wide range of Chinese NLP tasks, including machine reading comprehension, natural language inference, text classification, sentence pair matching, and competitive performances in named entity recognition. Code and pretrained models are publicly available at https://github.com/ShannonAI/ChineseBert.

Yongchun Zhu, Fuzhen Zhuang, Jindong Wang et al.

For a target task where labeled data is unavailable, domain adaptation can transfer a learner from a different source domain. Previous deep domain adaptation methods mainly learn a global domain shift, i.e., align the global source and target distributions without considering the relationships between two subdomains within the same category of different domains, leading to unsatisfying transfer learning performance without capturing the fine-grained information. Recently, more and more researchers pay attention to Subdomain Adaptation which focuses on accurately aligning the distributions of the relevant subdomains. However, most of them are adversarial methods which contain several loss functions and converge slowly. Based on this, we present Deep Subdomain Adaptation Network (DSAN) which learns a transfer network by aligning the relevant subdomain distributions of domain-specific layer activations across different domains based on a local maximum mean discrepancy (LMMD). Our DSAN is very simple but effective which does not need adversarial training and converges fast. The adaptation can be achieved easily with most feed-forward network models by extending them with LMMD loss, which can be trained efficiently via back-propagation. Experiments demonstrate that DSAN can achieve remarkable results on both object recognition tasks and digit classification tasks. Our code will be available at: https://github.com/easezyc/deep-transfer-learning

Fuzhen Zhuang, Keyu Duan, Tongjia Guo et al.

The transfer learning toolkit wraps the codes of 17 transfer learning models and provides integrated interfaces, allowing users to use those models by calling a simple function. It is easy for primary researchers to use this toolkit and to choose proper models for real-world applications. The toolkit is written in Python and distributed under MIT open source license. In this paper, the current state of this toolkit is described and the necessary environment setting and usage are introduced.

Zhilong Zhang, Lei Zhang, Qing He et al.

Open world object detection faces a significant challenge in domain-invariant representation, i.e., implicit non-causal factors. Most domain generalization (DG) methods based on domain adversarial learning (DAL) pay much attention to learn domain-invariant information, but often overlook the potential non-causal factors. We unveil two critical causes: 1) The domain discriminator-based DAL method is subject to the extremely sparse domain label, i.e., assigning only one domain label to each dataset, thus can only associate explicit non-causal factor, which is incredibly limited. 2) The non-causal factors, induced by unidentified data bias, are excessively implicit and cannot be solely discerned by conventional DAL paradigm. Based on these key findings, inspired by the Granular-Ball perspective, we propose an improved DAL method, i.e., GB-DAL. The proposed GB-DAL utilizes Prototype-based Granular Ball Splitting (PGBS) module to generate more dense domains from limited datasets, akin to more fine-grained granular balls, indicating more potential non-causal factors. Inspired by adversarial perturbations akin to non-causal factors, we propose a Simulated Non-causal Factors (SNF) module as a means of data augmentation to reduce the implicitness of non-causal factors, and facilitate the training of GB-DAL. Comparative experiments on numerous benchmarks demonstrate that our method achieves better generalization performance in novel circumstances.

Hanyu Zhang, Xiting Wang, Chengao Li et al.

As large language models (LLMs) are widely deployed across various domains, the ability to control their generated outputs has become more critical. This control involves aligning LLMs outputs with human values and ethical principles or customizing LLMs on specific topics or styles for individual users. Existing controlled generation methods either require significant computational resources and extensive trial-and-error or provide coarse-grained control. In this paper, we propose Generation with Concept Activation Vector (GCAV), a lightweight model control framework that ensures accurate control without requiring resource-extensive fine-tuning. Specifically, GCAV first trains a concept activation vector for specified concepts to be controlled, such as toxicity. During inference, GCAV steers the concept vector in LLMs, for example, by removing the toxicity concept vector from the activation layers. Control experiments from different perspectives, including toxicity reduction, sentiment control, linguistic style, and topic control, demonstrate that our framework achieves state-of-the-art performance with granular control, allowing for fine-grained adjustments of both the steering layers and the steering magnitudes for individual samples.

Maohao Shen, Shun Zhang, Jilong Wu et al.

Large language models (LLMs) have revolutionized natural language processing (NLP) with impressive performance across various text-based tasks. However, the extension of text-dominant LLMs to with speech generation tasks remains under-explored. In this work, we introduce a text-to-speech (TTS) system powered by a fine-tuned Llama model, named TTS-Llama, that achieves state-of-the-art speech synthesis performance. Building on TTS-Llama, we further propose MoLE-Llama, a text-and-speech multimodal LLM developed through purely late-fusion parameter-efficient fine-tuning (PEFT) and a mixture-of-expert architecture. Extensive empirical results demonstrate MoLE-Llama's competitive performance on both text-only question-answering (QA) and TTS tasks, mitigating catastrophic forgetting issue in either modality. Finally, we further explore MoLE-Llama in text-in-speech-out QA tasks, demonstrating its great potential as a multimodal dialog system capable of speech generation.

Tianyu Chen, Yiming Zhang, Guoxin Yu et al.

In this paper, we extend financial sentiment analysis~(FSA) to event-level since events usually serve as the subject of the sentiment in financial text. Though extracting events from the financial text may be conducive to accurate sentiment predictions, it has specialized challenges due to the lengthy and discontinuity of events in a financial text. To this end, we reconceptualize the event extraction as a classification task by designing a categorization comprising coarse-grained and fine-grained event categories. Under this setting, we formulate the \textbf{E}vent-Level \textbf{F}inancial \textbf{S}entiment \textbf{A}nalysis~(\textbf{EFSA} for short) task that outputs quintuples consisting of (company, industry, coarse-grained event, fine-grained event, sentiment) from financial text. A large-scale Chinese dataset containing $12,160$ news articles and $13,725$ quintuples is publicized as a brand new testbed for our task. A four-hop Chain-of-Thought LLM-based approach is devised for this task. Systematically investigations are conducted on our dataset, and the empirical results demonstrate the benchmarking scores of existing methods and our proposed method can reach the current state-of-the-art. Our dataset and framework implementation are available at https://anonymous.4open.science/r/EFSA-645E

Chengao Li, Hanyu Zhang, Yunkun Xu et al.

Reinforcement Learning from Human Feedback (RLHF) has emerged as a powerful technique for aligning large language models (LLMs) with human preferences. However, effectively aligning LLMs with diverse human preferences remains a significant challenge, particularly when they are conflict. To address this issue, we frame human value alignment as a multi-objective optimization problem, aiming to maximize a set of potentially conflicting objectives. We introduce Gradient-Adaptive Policy Optimization (GAPO), a novel fine-tuning paradigm that employs multiple-gradient descent to align LLMs with diverse preference distributions. GAPO adaptively rescales the gradients for each objective to determine an update direction that optimally balances the trade-offs between objectives. Additionally, we introduce P-GAPO, which incorporates user preferences across different objectives and achieves Pareto solutions that better align with the user's specific needs. Our theoretical analysis demonstrates that GAPO converges towards a Pareto optimal solution for multiple objectives. Empirical results on Mistral-7B show that GAPO outperforms current state-of-the-art methods, achieving superior performance in both helpfulness and harmlessness.

Junhong Lian, Xiang Ao, Xinyu Liu et al.

Personalized news headline generation aims to provide users with attention-grabbing headlines that are tailored to their preferences. Prevailing methods focus on user-oriented content preferences, but most of them overlook the fact that diverse stylistic preferences are integral to users' panoramic interests, leading to suboptimal personalization. In view of this, we propose a novel Stylistic-Content Aware Personalized Headline Generation (SCAPE) framework. SCAPE extracts both content and stylistic features from headlines with the aid of large language model (LLM) collaboration. It further adaptively integrates users' long- and short-term interests through a contrastive learning-based hierarchical fusion network. By incorporating the panoramic interests into the headline generator, SCAPE reflects users' stylistic-content preferences during the generation process. Extensive experiments on the real-world dataset PENS demonstrate the superiority of SCAPE over baselines.

Wonjune Kang, Yun Wang, Shun Zhang et al. · mit

We propose a multi-task learning (MTL) model for jointly performing three tasks that are commonly solved in a text-to-speech (TTS) front-end: text normalization (TN), part-of-speech (POS) tagging, and homograph disambiguation (HD). Our framework utilizes a tree-like structure with a trunk that learns shared representations, followed by separate task-specific heads. We further incorporate a pre-trained language model to utilize its built-in lexical and contextual knowledge, and study how to best use its embeddings so as to most effectively benefit our multi-task model. Through task-wise ablations, we show that our full model trained on all three tasks achieves the strongest overall performance compared to models trained on individual or sub-combinations of tasks, confirming the advantages of our MTL framework. Finally, we introduce a new HD dataset containing a balanced number of sentences in diverse contexts for a variety of homographs and their pronunciations. We demonstrate that incorporating this dataset into training significantly improves HD performance over only using a commonly used, but imbalanced, pre-existing dataset.

Cuihui Xia, Lei Yue, Deliang Chen et al.

Traditional equation-driven hydrological models often struggle to accurately predict streamflow in challenging regional Earth systems like the Tibetan Plateau, while hybrid and existing algorithm-driven models face difficulties in interpreting hydrological behaviors. This work introduces HydroTrace, an algorithm-driven, data-agnostic model that substantially outperforms these approaches, achieving a Nash-Sutcliffe Efficiency of 98% and demonstrating strong generalization on unseen data. Moreover, HydroTrace leverages advanced attention mechanisms to capture spatial-temporal variations and feature-specific impacts, enabling the quantification and spatial resolution of streamflow partitioning as well as the interpretation of hydrological behaviors such as glacier-snow-streamflow interactions and monsoon dynamics. Additionally, a large language model (LLM)-based application allows users to easily understand and apply HydroTrace's insights for practical purposes. These advancements position HydroTrace as a transformative tool in hydrological and broader Earth system modeling, offering enhanced prediction accuracy and interpretability.

Chenyang Wang, Roger Olsson, Stefan Forsström et al.

Empowered by deep learning, semantic communication marks a paradigm shift from transmitting raw data to conveying task-relevant meaning, enabling more efficient and intelligent wireless systems. In this study, we explore a deep learning-based task-oriented communication framework that jointly considers classification performance, computational latency, and communication cost. We adopt ResNets-based models and evaluate them on the CIFAR-10 and CIFAR-100 datasets to simulate real-world classification tasks in wireless environments. We partition the model at various points to simulate split inference across a wireless channel. By varying the split location and the size of the transmitted semantic feature vector, we systematically analyze the trade-offs between task accuracy and resource efficiency. Experimental results show that, with appropriate model partitioning and semantic feature compression, the system can retain over 85\% of baseline accuracy while significantly reducing both computational load and communication overhead.

Qing He, Wei Cai

We propose a deep neural network architecture designed such that its output forms an invertible symplectomorphism of the input. This design draws an analogy to the real-valued non-volume-preserving (real NVP) method used in normalizing flow techniques. Utilizing this neural network type allows for learning tasks on unknown Hamiltonian systems without breaking the inherent symplectic structure of the phase space.

Prabhav Agrawal, Thilo Koehler, Zhiping Xiu et al.

Neural vocoders model the raw audio waveform and synthesize high-quality audio, but even the highly efficient ones, like MB-MelGAN and LPCNet, fail to run real-time on a low-end device like a smartglass. A pure digital signal processing (DSP) based vocoder can be implemented via lightweight fast Fourier transforms (FFT), and therefore, is a magnitude faster than any neural vocoder. A DSP vocoder often gets a lower audio quality due to consuming over-smoothed acoustic model predictions of approximate representations for the vocal tract. In this paper, we propose an ultra-lightweight differential DSP (DDSP) vocoder that uses a jointly optimized acoustic model with a DSP vocoder, and learns without an extracted spectral feature for the vocal tract. The model achieves audio quality comparable to neural vocoders with a high average MOS of 4.36 while being efficient as a DSP vocoder. Our C++ implementation, without any hardware-specific optimization, is at 15 MFLOPS, surpasses MB-MelGAN by 340 times in terms of FLOPS, and achieves a vocoder-only RTF of 0.003 and overall RTF of 0.044 while running single-threaded on a 2GHz Intel Xeon CPU.

Shuo Yu, Hongyan Xue, Xiang Ao et al.

In the field of quantitative trading, it is common practice to transform raw historical stock data into indicative signals for the market trend. Such signals are called alpha factors. Alphas in formula forms are more interpretable and thus favored by practitioners concerned with risk. In practice, a set of formulaic alphas is often used together for better modeling precision, so we need to find synergistic formulaic alpha sets that work well together. However, most traditional alpha generators mine alphas one by one separately, overlooking the fact that the alphas would be combined later. In this paper, we propose a new alpha-mining framework that prioritizes mining a synergistic set of alphas, i.e., it directly uses the performance of the downstream combination model to optimize the alpha generator. Our framework also leverages the strong exploratory capabilities of reinforcement learning~(RL) to better explore the vast search space of formulaic alphas. The contribution to the combination models' performance is assigned to be the return used in the RL process, driving the alpha generator to find better alphas that improve upon the current set. Experimental evaluations on real-world stock market data demonstrate both the effectiveness and the efficiency of our framework for stock trend forecasting. The investment simulation results show that our framework is able to achieve higher returns compared to previous approaches.

Yongchun Zhu, Dongbo Xi, Bowen Song et al.

With the explosive growth of the e-commerce industry, detecting online transaction fraud in real-world applications has become increasingly important to the development of e-commerce platforms. The sequential behavior history of users provides useful information in differentiating fraudulent payments from regular ones. Recently, some approaches have been proposed to solve this sequence-based fraud detection problem. However, these methods usually suffer from two problems: the prediction results are difficult to explain and the exploitation of the internal information of behaviors is insufficient. To tackle the above two problems, we propose a Hierarchical Explainable Network (HEN) to model users' behavior sequences, which could not only improve the performance of fraud detection but also make the inference process interpretable. Meanwhile, as e-commerce business expands to new domains, e.g., new countries or new markets, one major problem for modeling user behavior in fraud detection systems is the limitation of data collection, e.g., very few data/labels available. Thus, in this paper, we further propose a transfer framework to tackle the cross-domain fraud detection problem, which aims to transfer knowledge from existing domains (source domains) with enough and mature data to improve the performance in the new domain (target domain). Our proposed method is a general transfer framework that could not only be applied upon HEN but also various existing models in the Embedding & MLP paradigm. Based on 90 transfer task experiments, we also demonstrate that our transfer framework could not only contribute to the cross-domain fraud detection task with HEN, but also be universal and expandable for various existing models.

Dongbo Xi, Fuzhen Zhuang, Yanchi Liu et al.

Recent years have witnessed the increasing popularity of Location-based Social Network (LBSN) services, which provides unparalleled opportunities to build personalized Point-of-Interest (POI) recommender systems. Existing POI recommendation and location prediction tasks utilize past information for future recommendation or prediction from a single direction perspective, while the missing POI category identification task needs to utilize the check-in information both before and after the missing category. Therefore, a long-standing challenge is how to effectively identify the missing POI categories at any time in the real-world check-in data of mobile users. To this end, in this paper, we propose a novel neural network approach to identify the missing POI categories by integrating both bi-directional global non-personal transition patterns and personal preferences of users. Specifically, we delicately design an attention matching cell to model how well the check-in category information matches their non-personal transition patterns and personal preferences. Finally, we evaluate our model on two real-world datasets, which clearly validate its effectiveness compared with the state-of-the-art baselines. Furthermore, our model can be naturally extended to address next POI category recommendation and prediction tasks with competitive performance.

Dongbo Xi, Fuzhen Zhuang, Bowen Song et al.

Many prediction tasks of real-world applications need to model multi-order feature interactions in user's event sequence for better detection performance. However, existing popular solutions usually suffer two key issues: 1) only focusing on feature interactions and failing to capture the sequence influence; 2) only focusing on sequence information, but ignoring internal feature relations of each event, thus failing to extract a better event representation. In this paper, we consider a two-level structure for capturing the hierarchical information over user's event sequence: 1) learning effective feature interactions based event representation; 2) modeling the sequence representation of user's historical events. Experimental results on both industrial and public datasets clearly demonstrate that our model achieves significantly better performance compared with state-of-the-art baselines.

Dongbo Xi, Fuzhen Zhuang, Ganbin Zhou et al.

Domain adaptation tasks such as cross-domain sentiment classification aim to utilize existing labeled data in the source domain and unlabeled or few labeled data in the target domain to improve the performance in the target domain via reducing the shift between the data distributions. Existing cross-domain sentiment classification methods need to distinguish pivots, i.e., the domain-shared sentiment words, and non-pivots, i.e., the domain-specific sentiment words, for excellent adaptation performance. In this paper, we first design a Category Attention Network (CAN), and then propose a model named CAN-CNN to integrate CAN and a Convolutional Neural Network (CNN). On the one hand, the model regards pivots and non-pivots as unified category attribute words and can automatically capture them to improve the domain adaptation performance; on the other hand, the model makes an attempt at interpretability to learn the transferred category attribute words. Specifically, the optimization objective of our model has three different components: 1) the supervised classification loss; 2) the distributions loss of category feature weights; 3) the domain invariance loss. Finally, the proposed model is evaluated on three public sentiment analysis datasets and the results demonstrate that CAN-CNN can outperform other various baseline methods.

Dongbo Xi, Fuzhen Zhuang, Yanchi Liu et al.

Human mobility data accumulated from Point-of-Interest (POI) check-ins provides great opportunity for user behavior understanding. However, data quality issues (e.g., geolocation information missing, unreal check-ins, data sparsity) in real-life mobility data limit the effectiveness of existing POI-oriented studies, e.g., POI recommendation and location prediction, when applied to real applications. To this end, in this paper, we develop a model, named Bi-STDDP, which can integrate bi-directional spatio-temporal dependence and users' dynamic preferences, to identify the missing POI check-in where a user has visited at a specific time. Specifically, we first utilize bi-directional global spatial and local temporal information of POIs to capture the complex dependence relationships. Then, target temporal pattern in combination with user and POI information are fed into a multi-layer network to capture users' dynamic preferences. Moreover, the dynamic preferences are transformed into the same space as the dependence relationships to form the final model. Finally, the proposed model is evaluated on three large-scale real-world datasets and the results demonstrate significant improvements of our model compared with state-of-the-art methods. Also, it is worth noting that the proposed model can be naturally extended to address POI recommendation and location prediction tasks with competitive performances.

Fuwei Zhang, Zhao Zhang, Xiang Ao et al.

Cross-Lingual Information Retrieval (CLIR) aims to rank the documents written in a language different from the user's query. The intrinsic gap between different languages is an essential challenge for CLIR. In this paper, we introduce the multilingual knowledge graph (KG) to the CLIR task due to the sufficient information of entities in multiple languages. It is regarded as a "silver bullet" to simultaneously perform explicit alignment between queries and documents and also broaden the representations of queries. And we propose a model named CLIR with hierarchical knowledge enhancement (HIKE) for our task. The proposed model encodes the textual information in queries, documents and the KG with multilingual BERT, and incorporates the KG information in the query-document matching process with a hierarchical information fusion mechanism. Particularly, HIKE first integrates the entities and their neighborhood in KG into query representations with a knowledge-level fusion, then combines the knowledge from both source and target languages to further mitigate the linguistic gap with a language-level fusion. Finally, experimental results demonstrate that HIKE achieves substantial improvements over state-of-the-art competitors.

Zhao Zhang, Fuzhen Zhuang, Hengshu Zhu et al.

Nowadays, Knowledge graphs (KGs) have been playing a pivotal role in AI-related applications. Despite the large sizes, existing KGs are far from complete and comprehensive. In order to continuously enrich KGs, automatic knowledge construction and update mechanisms are usually utilized, which inevitably bring in plenty of noise. However, most existing knowledge graph embedding (KGE) methods assume that all the triple facts in KGs are correct, and project both entities and relations into a low-dimensional space without considering noise and knowledge conflicts. This will lead to low-quality and unreliable representations of KGs. To this end, in this paper, we propose a general multi-task reinforcement learning framework, which can greatly alleviate the noisy data problem. In our framework, we exploit reinforcement learning for choosing high-quality knowledge triples while filtering out the noisy ones. Also, in order to take full advantage of the correlations among semantically similar relations, the triple selection processes of similar relations are trained in a collective way with multi-task learning. Moreover, we extend popular KGE models TransE, DistMult, ConvE and RotatE with the proposed framework. Finally, the experimental validation shows that our approach is able to enhance existing KGE models and can provide more robust representations of KGs in noisy scenarios.

Yuxian Meng, Xiang Ao, Qing He et al.

A long-standing issue with paraphrase generation is how to obtain reliable supervision signals. In this paper, we propose an unsupervised paradigm for paraphrase generation based on the assumption that the probabilities of generating two sentences with the same meaning given the same context should be the same. Inspired by this fundamental idea, we propose a pipelined system which consists of paraphrase candidate generation based on contextual language models, candidate filtering using scoring functions, and paraphrase model training based on the selected candidates. The proposed paradigm offers merits over existing paraphrase generation methods: (1) using the context regularizer on meanings, the model is able to generate massive amounts of high-quality paraphrase pairs; and (2) using human-interpretable scoring functions to select paraphrase pairs from candidates, the proposed framework provides a channel for developers to intervene with the data generation process, leading to a more controllable model. Experimental results across different tasks and datasets demonstrate that the effectiveness of the proposed model in both supervised and unsupervised setups.

Haoming Li, Feiyang Pan, Xiang Ao et al.

The delayed feedback problem is one of the imperative challenges in online advertising, which is caused by the highly diversified feedback delay of a conversion varying from a few minutes to several days. It is hard to design an appropriate online learning system under these non-identical delay for different types of ads and users. In this paper, we propose to tackle the delayed feedback problem in online advertising by "Following the Prophet" (FTP for short). The key insight is that, if the feedback came instantly for all the logged samples, we could get a model without delayed feedback, namely the "prophet". Although the prophet cannot be obtained during online learning, we show that we could predict the prophet's predictions by an aggregation policy on top of a set of multi-task predictions, where each task captures the feedback patterns of different periods. We propose the objective and optimization approach for the policy, and use the logged data to imitate the prophet. Extensive experiments on three real-world advertising datasets show that our method outperforms the previous state-of-the-art baselines.

Feiyang Pan, Haoming Li, Xiang Ao et al.

The exploration/exploitation (E&E) dilemma lies at the core of interactive systems such as online advertising, for which contextual bandit algorithms have been proposed. Bayesian approaches provide guided exploration with principled uncertainty estimation, but the applicability is often limited due to over-simplified assumptions. Non-Bayesian bootstrap methods, on the other hand, can apply to complex problems by using deep reward models, but lacks clear guidance to the exploration behavior. It still remains largely unsolved to develop a practical method for complex deep contextual bandits. In this paper, we introduce Guided Bootstrap (GuideBoot for short), combining the best of both worlds. GuideBoot provides explicit guidance to the exploration behavior by training multiple models over both real samples and noisy samples with fake labels, where the noise is added according to the predictive uncertainty. The proposed method is efficient as it can make decisions on-the-fly by utilizing only one randomly chosen model, but is also effective as we show that it can be viewed as a non-Bayesian approximation of Thompson sampling. Moreover, we extend it to an online version that can learn solely from streaming data, which is favored in real applications. Extensive experiments on both synthetic task and large-scale advertising environments show that GuideBoot achieves significant improvements against previous state-of-the-art methods.

Ann Lee, Peng-Jen Chen, Changhan Wang et al.

We present a direct speech-to-speech translation (S2ST) model that translates speech from one language to speech in another language without relying on intermediate text generation. We tackle the problem by first applying a self-supervised discrete speech encoder on the target speech and then training a sequence-to-sequence speech-to-unit translation (S2UT) model to predict the discrete representations of the target speech. When target text transcripts are available, we design a joint speech and text training framework that enables the model to generate dual modality output (speech and text) simultaneously in the same inference pass. Experiments on the Fisher Spanish-English dataset show that the proposed framework yields improvement of 6.7 BLEU compared with a baseline direct S2ST model that predicts spectrogram features. When trained without any text transcripts, our model performance is comparable to models that predict spectrograms and are trained with text supervision, showing the potential of our system for translation between unwritten languages. Audio samples are available at https://facebookresearch.github.io/speech_translation/direct_s2st_units/index.html .

Hao Chen, Fuzhen Zhuang, Li Xiao et al.

Recently, Graph Convolutional Networks (GCNs) have proven to be a powerful mean for Computer Aided Diagnosis (CADx). This approach requires building a population graph to aggregate structural information, where the graph adjacency matrix represents the relationship between nodes. Until now, this adjacency matrix is usually defined manually based on phenotypic information. In this paper, we propose an encoder that automatically selects the appropriate phenotypic measures according to their spatial distribution, and uses the text similarity awareness mechanism to calculate the edge weights between nodes. The encoder can automatically construct the population graph using phenotypic measures which have a positive impact on the final results, and further realizes the fusion of multimodal information. In addition, a novel graph convolution network architecture using multi-layer aggregation mechanism is proposed. The structure can obtain deep structure information while suppressing over-smooth, and increase the similarity between the same type of nodes. Experimental results on two databases show that our method can significantly improve the diagnostic accuracy for Autism spectrum disorder and breast cancer, indicating its universality in leveraging multimodal data for disease prediction.

Yongchun Zhu, Kaikai Ge, Fuzhen Zhuang et al.

Cold-start problems are enormous challenges in practical recommender systems. One promising solution for this problem is cross-domain recommendation (CDR) which leverages rich information from an auxiliary (source) domain to improve the performance of recommender system in the target domain. In these CDR approaches, the family of Embedding and Mapping methods for CDR (EMCDR) is very effective, which explicitly learn a mapping function from source embeddings to target embeddings with overlapping users. However, these approaches suffer from one serious problem: the mapping function is only learned on limited overlapping users, and the function would be biased to the limited overlapping users, which leads to unsatisfying generalization ability and degrades the performance on cold-start users in the target domain. With the advantage of meta learning which has good generalization ability to novel tasks, we propose a transfer-meta framework for CDR (TMCDR) which has a transfer stage and a meta stage. In the transfer (pre-training) stage, a source model and a target model are trained on source and target domains, respectively. In the meta stage, a task-oriented meta network is learned to implicitly transform the user embedding in the source domain to the target feature space. In addition, the TMCDR is a general framework that can be applied upon various base models, e.g., MF, BPR, CML. By utilizing data from Amazon and Douban, we conduct extensive experiments on 6 cross-domain tasks to demonstrate the superior performance and compatibility of TMCDR.

Qing He, Zhiping Xiu, Thilo Koehler et al.

Typical high quality text-to-speech (TTS) systems today use a two-stage architecture, with a spectrum model stage that generates spectral frames and a vocoder stage that generates the actual audio. High-quality spectrum models usually incorporate the encoder-decoder architecture with self-attention or bi-directional long short-term (BLSTM) units. While these models can produce high quality speech, they often incur O($L$) increase in both latency and real-time factor (RTF) with respect to input length $L$. In other words, longer inputs leads to longer delay and slower synthesis speed, limiting its use in real-time applications. In this paper, we propose a multi-rate attention architecture that breaks the latency and RTF bottlenecks by computing a compact representation during encoding and recurrently generating the attention vector in a streaming manner during decoding. The proposed architecture achieves high audio quality (MOS of 4.31 compared to groundtruth 4.48), low latency, and low RTF at the same time. Meanwhile, both latency and RTF of the proposed system stay constant regardless of input lengths, making it ideal for real-time applications.

Yongchun Zhu, Fuzhen Zhuang, Xiangliang Zhang et al.

Many few-shot learning approaches have been designed under the meta-learning framework, which learns from a variety of learning tasks and generalizes to new tasks. These meta-learning approaches achieve the expected performance in the scenario where all samples are drawn from the same distributions (i.i.d. observations). However, in real-world applications, few-shot learning paradigm often suffers from data shift, i.e., samples in different tasks, even in the same task, could be drawn from various data distributions. Most existing few-shot learning approaches are not designed with the consideration of data shift, and thus show downgraded performance when data distribution shifts. However, it is non-trivial to address the data shift problem in few-shot learning, due to the limited number of labeled samples in each task. Targeting at addressing this problem, we propose a novel metric-based meta-learning framework to extract task-specific representations and task-shared representations with the help of knowledge graph. The data shift within/between tasks can thus be combated by the combination of task-shared and task-specific representations. The proposed model is evaluated on popular benchmarks and two constructed new challenging datasets. The evaluation results demonstrate its remarkable performance.