Zihao Chen

Zihao Chen

Anchored fixed point and monotone equation methods, including Halpern iteration, extra anchored gradient, and their relatives, add a vanishing pull toward a reference point to obtain last-iterate guarantees. Existing anchored variants often achieve sharp last-iterate guarantees, but from the update-level perspective the placement of the anchor can be algorithm-specific and conceptually opaque. We show that anchoring admits a single operator-side construction: regularize the operator queried by the base method with a vanishing Tikhonov term, then run the unmodified base method. Applied to the Picard iteration, this recipe reproduces the Halpern iteration; applied to the forward step, extragradient (EG), and past extragradient (PEG, also known as Popov's method), it yields three variants whose anchor placements inherit the base method's query pattern. The forward-step instantiation gives a new residual convergence guarantee, while the EG and PEG instantiations give new regularized variants. The four analyses share a residual recurrence, recovering the $O(1/k)$ Halpern residual-norm convergence rate, giving $O(1/\sqrt{k})$ for the regularized forward step, and giving $O(1/k)$ for the regularized EG and PEG variants in the unconstrained monotone Lipschitz setting.

Junjie Zheng, Huixin Xue, Shihong Ren et al.

Text-to-speech (TTS) and singing voice synthesis (SVS) both aim to generate human vocal audio from symbolic inputs, but they impose different requirements on the generation process. Speech generation relies on flexible, language-driven prosody, whereas singing generation requires explicit melody control and accurate rhythmic alignment. This mismatch makes it challenging to train a single model that can generate both natural speech and controllable singing, since melody-related conditions should strongly constrain singing but should not restrict speech prosody. We present UniVoice, a unified speech and singing voice generation framework based on conditional flow matching. Instead of using a single undifferentiated conditioning representation, UniVoice factorizes the condition into content, melody, and timbre, which are encoded by modality-appropriate encoders and consumed by a shared Diffusion Transformer (DiT) backbone. For singing, the melody condition is represented by MIDI note sequences; for speech, it is replaced with a learned null melody token, allowing the model to infer prosody from linguistic and acoustic context. This design preserves explicit melody control for singing while avoiding the need to impose melody constraints on speech. We further analyze the null melody token as an approximation to melody marginalization in the conditional flow. Trained on 30k hours of speech and 35k hours of singing data, UniVoice achieves a speech PER of 5.26\%, comparable to dedicated TTS systems such as F5-TTS (5.21\%) and CosyVoice3 (5.30\%). On singing generation, UniVoice achieves a PER of 16.22\%, outperforming the unified baseline Vevo1.5 (24.72\%).

Carole H. Sudre, Kimberlin Van Wijnen, Florian Dubost et al.

Imaging markers of cerebral small vessel disease provide valuable information on brain health, but their manual assessment is time-consuming and hampered by substantial intra- and interrater variability. Automated rating may benefit biomedical research, as well as clinical assessment, but diagnostic reliability of existing algorithms is unknown. Here, we present the results of the \textit{VAscular Lesions DetectiOn and Segmentation} (\textit{Where is VALDO?}) challenge that was run as a satellite event at the international conference on Medical Image Computing and Computer Aided Intervention (MICCAI) 2021. This challenge aimed to promote the development of methods for automated detection and segmentation of small and sparse imaging markers of cerebral small vessel disease, namely enlarged perivascular spaces (EPVS) (Task 1), cerebral microbleeds (Task 2) and lacunes of presumed vascular origin (Task 3) while leveraging weak and noisy labels. Overall, 12 teams participated in the challenge proposing solutions for one or more tasks (4 for Task 1 - EPVS, 9 for Task 2 - Microbleeds and 6 for Task 3 - Lacunes). Multi-cohort data was used in both training and evaluation. Results showed a large variability in performance both across teams and across tasks, with promising results notably for Task 1 - EPVS and Task 2 - Microbleeds and not practically useful results yet for Task 3 - Lacunes. It also highlighted the performance inconsistency across cases that may deter use at an individual level, while still proving useful at a population level.

Zihao Chen, Yuhua Chen, Yibin Xie et al.

Non-Cartesian sampling with subspace-constrained image reconstruction is a popular approach to dynamic MRI, but slow iterative reconstruction limits its clinical application. Data-consistent (DC) deep learning can accelerate reconstruction with good image quality, but has not been formulated for non-Cartesian subspace imaging. In this study, we propose a DC non-Cartesian deep subspace learning framework for fast, accurate dynamic MR image reconstruction. Four novel DC formulations are developed and evaluated: two gradient decent approaches, a directly solved approach, and a conjugate gradient approach. We applied a U-Net model with and without DC layers to reconstruct T1-weighted images for cardiac MR Multitasking (an advanced multidimensional imaging method), comparing our results to the iteratively reconstructed reference. Experimental results show that the proposed framework significantly improves reconstruction accuracy over the U-Net model without DC, while significantly accelerating the reconstruction over conventional iterative reconstruction.

Ya-Qi Yu, Hao Wang, Fangyu Hong et al.

While Reinforcement Learning with Verifiable Rewards (RLVR) is effective for deterministically checkable tasks, many vision-language tasks are partially verifiable, demanding multi-criteria supervision (e.g., perceptual details, reasoning steps, and constraints). Rubrics provide a natural interface for this fine-grained supervision, but their effectiveness depends on the execution accuracy during online RL. We propose Reinforcement Learning with Robust Rubric Rewards ($\text{RLR}^3$), extending RLVR from task-level verification to criterion-level verification. $\text{RLR}^3$ routes instance-specific rubrics through two execution paths: an LLM-as-an-extractor paired with a deterministic verifier, or an LLM-as-a-Judge for non-verifiable criteria. To ensure faithful scoring, $\text{RLR}^3$ introduce a minimal exposure strategy that masks ground truths from extractors and images from judges. Furthermore, $\text{RLR}^3$ employs hierarchical aggregation to prioritize essential criteria over additional criteria, and mitigates score saturation within rollout groups. Evaluated on Qwen3-VL-30B-A3B across 15 benchmarks, $\text{RLR}^3$ consistently outperforms RLVR, yielding a 4.7-point improvement over the base model and exceeding the official instruct-to-thinking model gap. Controlled audits confirm our deterministic verification and minimal exposure significantly reduce exploitable false positives.

Zihao Chen, Jiayin Wang, Ziyi Sun et al.

This brief proposes \emph{White-Op}, an interpretable operational amplifier (op-amp) parameter design framework based on the human-mimicking reasoning of large-language-model agents. We formalize the implicit human reasoning mechanism into explicit steps of \emph{\textbf{introducing hypothetical constraints}}, and develop an iterative, human-like \emph{\textbf{hypothesis-verification-decision}} workflow. Specifically, the agent is guided to introduce hypothetical constraints to derive and properly regulate positions of symbolically tractable poles and zeros, thus formulating a closed-form mathematical optimization problem, which is then solved programmatically and verified via simulation. Theory-simulation result analysis guides the decision-making for refinement. Experiments on 9 op-amp topologies show that, unlike the uninterpretable black-box baseline which finally fails in 5 topologies, White-Op achieves reliable, interpretable behavioral-level designs with only 8.52\% theoretical prediction error and the design functionality retains after transistor-level mapping for all topologies. White-Op is open-sourced at \textcolor{blue}{https://github.com/zhchenfdu/whiteop}.

Zihao Chen, Alexandre Andre, Wenrui Ma et al.

Forecasting is critical in areas such as finance, biology, and healthcare. Despite the progress in the field, making accurate forecasts remains challenging because real-world time series contain both global trends, local fine-grained structure, and features on multiple scales in between. Here, we present a new forecasting method, PRISM (Partitioned Representation for Iterative Sequence Modeling), that addresses this challenge through a learnable tree-based partitioning of the signal. At the root of the tree, a global representation captures coarse trends in the signal, while recursive splits reveal increasingly localized views of the signal. At each level of the tree, data are projected onto a time-frequency basis (e.g., wavelets or exponential moving averages) to extract scale-specific features, which are then aggregated across the hierarchy. This design allows the model to jointly capture global structure and local dynamics of the signal, enabling accurate forecasting. Experiments across benchmark datasets show that our method outperforms state-of-the-art methods for forecasting. Overall, these results demonstrate that our hierarchical approach provides a lightweight and flexible framework for forecasting multivariate time series. The code is available at https://github.com/nerdslab/prism.

Likang Wu, Zihao Chen, Jianxin Zhang et al.

With the rapid emergence of multi-behavior learning in recommender systems, leveraging auxiliary user behaviors has proven effective for mitigating target-behavior data sparsity. Yet auxiliary behavior graphs frequently contain noisy or irrelevant interactions that do not align with the target task, impeding the learning of accurate user and item embeddings. Moreover, the scarcity of direct supervision from the target behavior complicates the extraction of informative collaborative signals. In this paper, we introduce GCIB (Graph Contrastive Information Bottleneck), a novel framework that denoises auxiliary behavior information and enriches target behavior representations at both the structural and feature levels. At the structural level, GCIB employs a Graph Information Bottleneck (GIB) objective to maximize mutual information between the denoised auxiliary graph and the target-behavior graph while minimizing mutual information with the original auxiliary graph. This formulation preserves task-relevant structural patterns and suppresses spurious interactions. At the feature level, we propose a cross-behavior Graph Contrastive Learning (GCL) scheme in which denoised auxiliary features and target-behavior features serve as complementary views for both users and items. By contrasting these views, GCIB enriches sparse target-behavior representations with semantics distilled from auxiliary behaviors. Extensive experiments demonstrate that GCIB outperforms state-of-the-art baselines, highlighting its ability to learn noise-resilient and target-aware representations for multi-behavior recommendation.

Chunbo Hao, Junjie Zheng, Guobin Ma et al.

Regenerating singing voices with altered lyrics while preserving melody consistency remains challenging, as existing methods either offer limited controllability or require laborious manual alignment. We propose YingMusic-Singer, a fully diffusion-based model enabling melody-controllable singing voice synthesis with flexible lyric manipulation. The model takes three inputs: an optional timbre reference, a melody-providing singing clip, and modified lyrics, without manual alignment. Trained with curriculum learning and Group Relative Policy Optimization, YingMusic-Singer achieves stronger melody preservation and lyric adherence than Vevo2, the most comparable baseline supporting melody control without manual alignment. We also introduce LyricEditBench, the first benchmark for melody-preserving lyric modification evaluation. The code, weights, benchmark, and demos are publicly available at https://github.com/ASLP-lab/YingMusic-Singer.

Gongyu Chen, Xiaoyu Zhang, Zhenqiang Weng et al.

Singing voice conversion (SVC) aims to render the target singer's timbre while preserving melody and lyrics. However, existing zero-shot SVC systems remain fragile in real songs due to harmony interference, F0 errors, and the lack of inductive biases for singing. We propose YingMusic-SVC, a robust zero-shot framework that unifies continuous pre-training, robust supervised fine-tuning, and Flow-GRPO reinforcement learning. Our model introduces a singing-trained RVC timbre shifter for timbre-content disentanglement, an F0-aware timbre adaptor for dynamic vocal expression, and an energy-balanced rectified flow matching loss to enhance high-frequency fidelity. Experiments on a graded multi-track benchmark show that YingMusic-SVC achieves consistent improvements over strong open-source baselines in timbre similarity, intelligibility, and perceptual naturalness, especially under accompanied and harmony-contaminated conditions, demonstrating its effectiveness for real-world SVC deployment.

Zihao Chen, Hisashi Handa, Kimiaki Shirahama

Contrastive learning is widely used for sentence representation learning. Despite this prevalence, most studies have focused exclusively on English and few concern domain adaptation for domain-specific downstream tasks, especially for low-resource languages like Japanese, which are characterized by insufficient target domain data and the lack of a proper training strategy. To overcome this, we propose a novel Japanese sentence representation framework, JCSE (derived from ``Contrastive learning of Sentence Embeddings for Japanese''), that creates training data by generating sentences and synthesizing them with sentences available in a target domain. Specifically, a pre-trained data generator is finetuned to a target domain using our collected corpus. It is then used to generate contradictory sentence pairs that are used in contrastive learning for adapting a Japanese language model to a specific task in the target domain. Another problem of Japanese sentence representation learning is the difficulty of evaluating existing embedding methods due to the lack of benchmark datasets. Thus, we establish a comprehensive Japanese Semantic Textual Similarity (STS) benchmark on which various embedding models are evaluated. Based on this benchmark result, multiple embedding methods are chosen and compared with JCSE on two domain-specific tasks, STS in a clinical domain and information retrieval in an educational domain. The results show that JCSE achieves significant performance improvement surpassing direct transfer and other training strategies. This empirically demonstrates JCSE's effectiveness and practicability for downstream tasks of a low-resource language.

Ya-Qi Yu, Fangyu Hong, Xiangyang Qu et al.

The effectiveness of Direct Preference Optimization (DPO) depends on preference data that reflect the quality differences that matter in multimodal tasks. Existing pipelines often rely on off-policy perturbations or coarse outcome-based signals, which are not well suited to fine-grained visual reasoning. We propose rDPO, a preference optimization framework based on instance-specific rubrics. For each image-instruction pair, we create a checklist-style rubric of essential and additional criteria to score responses from any possible policies. The instruction-rubric pool is built offline and reused during the construction of on-policy data. On public reward modeling benchmarks, rubric-based prompting massively improves a 30B-A3B judge and brings it close to GPT-5.4. On public downstream benchmarks, rubric-based filtering raises the macro average to 82.69, whereas outcome-based filtering drops it to 75.82 from 81.14. When evaluating scalability on a comprehensive benchmark, rDPO achieves 61.01, markedly outperforming the style-constrained baseline (52.36) and surpassing the 59.48 base model. Together, these results show that visual preference optimization benefits from combining on-policy data construction with instance-specific criterion-level feedback.

Zihao Chen, Wenyong Wang, Sai Zou

In the absence of artificial labels, the independent and dependent features in the data are cluttered. How to construct the inductive biases of the model to flexibly divide and effectively contain features with different complexity is the main focal point of unsupervised disentangled representation learning. This paper proposes a new iterative decomposition path of total correlation and explains the disentangled representation ability of VAE from the perspective of model capacity allocation. The newly developed objective function combines latent variable dimensions into joint distribution while relieving the independence constraints of marginal distributions in combination, leading to latent variables with a more manipulable prior distribution. The novel model enables VAE to adjust the parameter capacity to divide dependent and independent data features flexibly. Experimental results on various datasets show an interesting relevance between model capacity and the latent variable grouping size, called the "V"-shaped best ELBO trajectory. Additionally, we empirically demonstrate that the proposed method obtains better disentangling performance with reasonable parameter capacity allocation.

Zihao Chen, Yeshwanth Cherapanamjeri

We investigate the quantitative performance of affine-equivariant estimators for robust mean estimation. As a natural stability requirement, the construction of such affine-equivariant estimators has been extensively studied in the statistics literature. We quantitatively evaluate these estimators under two outlier models which have been the subject of much recent work: the heavy-tailed and adversarial corruption settings. We establish lower bounds which show that affine-equivariance induces a strict degradation in recovery error with quantitative rates degrading by a factor of $\sqrt{d}$ in both settings. We find that classical estimators such as the Tukey median (Tukey '75) and Stahel-Donoho estimator (Stahel '81 and Donoho '82) are either quantitatively sub-optimal even within the class of affine-equivariant estimators or lack any quantitative guarantees. On the other hand, recent estimators with strong quantitative guarantees are not affine-equivariant or require additional distributional assumptions to achieve it. We remedy this by constructing a new affine-equivariant estimator which nearly matches our lower bound. Our estimator is based on a novel notion of a high-dimensional median which may be of independent interest. Notably, our results are applicable more broadly to any estimator whose performance is evaluated in the Mahalanobis norm which, for affine-equivariant estimators, corresponds to an evaluation in Euclidean norm on isotropic distributions.

Ziyan Lin, Zihao Chen

Magnetic Resonance Imaging (MRI) is important in clinic to produce high resolution images for diagnosis, but its acquisition time is long for high resolution images. Deep learning based MRI super resolution methods can reduce scan time without complicated sequence programming, but may create additional artifacts due to the discrepancy between training data and testing data. Data consistency layer can improve the deep learning results but needs raw k-space data. In this work, we propose a magnitude-image based data consistency deep learning MRI super resolution method to improve super resolution images' quality without raw k-space data. Our experiments show that the proposed method can improve NRMSE and SSIM of super resolution images compared to the same Convolutional Neural Network (CNN) block without data consistency module.

Zihao Chen, Wenyong Wang, Yu Xiang

Manifold learning has been proven to be an effective method for capturing the implicitly intrinsic structure of non-Euclidean data, in which one of the primary challenges is how to maintain the distortion-free (isometry) of the data representations. Actually, there is still no manifold learning method that provides a theoretical guarantee of isometry. Inspired by Nash's isometric theorem, we introduce a new concept called isometric immersion learning based on Riemannian geometry principles. Following this concept, an unsupervised neural network-based model that simultaneously achieves metric and manifold learning is proposed by integrating Riemannian geometry priors. What's more, we theoretically derive and algorithmically implement a maximum likelihood estimation-based training method for the new model. In the simulation experiments, we compared the new model with the state-of-the-art baselines on various 3-D geometry datasets, demonstrating that the new model exhibited significantly superior performance in multiple evaluation metrics. Moreover, we applied the Riemannian metric learned from the new model to downstream prediction tasks in real-world scenarios, and the accuracy was improved by an average of 8.8%.

Liangyi Huang, Zichen Liu, Fei Shao et al.

Security knowledge graphs can provide computable external memory for security agents, but constructing them from long-form cyber threat intelligence (CTI) remains difficult: LLMs often lack grounded security-domain knowledge, and end-to-end document-to-graph training is hard to supervise with cheap, stable rewards. We present GRID (Graph Representation of Intelligence Data), an end-to-end framework for security text knowledge graph construction. GRID first builds security-domain supervision from CTI articles by creating traceable article-graph alignments through graph extraction and knowledge-graph-conditioned text revision. It then turns document-to-graph learning into a scripted task bank combining four-option multi-select questions with triple-level regex matching targets, yielding more stable task-specific rewards than repeatedly scoring full graph outputs with an LLM judge. Using this supervision pipeline, we train two Qwen3-4B-Instruct-2507-based 4B extractors: a primary Task-bank Reward model and a secondary End2End Reward model with LLM-as-judge precision/recall rewards. On 249 CTI articles from GRID, CASIE, CTINexus, MalKG, and SecureNLP, the Task-bank Reward model with the ontology-guided GRID extraction pipeline reaches 84.62% source-averaged precision, 64.91% source-averaged recall, and 68.53% Avg F1, achieving the best source-averaged recall and near-top Avg F1 with lower token usage and deployment cost. The End2End Reward model reaches 76.91% precision, 53.85% recall, and 58.06% Avg F1. Further analyses show that task-bank rewards can be built once offline and reused across later post-training runs, outperforming online End2End LLM-as-judge reward and weaker alternatives such as Choice-only Reward and End2End SFT without RL.

Zihao Chen, Federico Botta

Travel time is a fundamental component of accessibility measurement, yet most accessibility analyses rely on static timetable data that assume public transport services operate exactly as scheduled. Such representations overlook the substantial variability in travel times arising from operational conditions and service disruptions. In this study, we develop a scalable framework for reconstructing empirical bus timetables from high-frequency vehicle location data. Using national-scale real-time feeds from the UK Bus Open Data Service (BODS), we implement an automated data collection pipeline that continuously archives vehicle positions and daily timetable data. Observed vehicle locations are then matched to scheduled routes to infer stop-level arrival and departure times, enabling the construction of corrected empirical timetables. The resulting dataset allows travel time variability (TTV) to be analysed at fine temporal resolution and across large geographic areas. The computational efficiency and scalability of the framework enable national-scale accessibility analyses that incorporate observed service performance, providing a more realistic evidence base for evaluating public transport services and supporting transport planning.

Zihao Chen, Xiaomeng Wang, Yuanjiang Huang et al.

The interpretability of model has become one of the obstacles to its wide application in the high-stake fields. The usual way to obtain interpretability is to build a black-box first and then explain it using the post-hoc methods. However, the explanations provided by the post-hoc method are not always reliable. Instead, we design an intrinsically interpretable model based on RRL(Rule Representation Learner) for the Lending Club dataset. Specifically, features can be divided into three categories according to their characteristics of themselves and build three sub-networks respectively, each of which is similar to a neural network with a single hidden layer but can be equivalently converted into a set of rules. During the training, we learned tricks from previous research to effectively train binary weights. Finally, our model is compared with the tree-based model. The results show that our model is much better than the interpretable decision tree in performance and close to other black-box, which is of practical significance to both financial institutions and borrowers. More importantly, our model is used to test the correctness of the explanations generated by the post-hoc method, the results show that the post-hoc method is not always reliable.

Hao Yan, Xingchen Liu, Hao Wang et al.

Recent strides in multimodal large language models (MLLMs) have significantly advanced their performance in many reasoning tasks. However, Abstract Visual Reasoning (AVR) remains a critical challenge, primarily due to limitations in perceiving abstract graphics. To tackle this issue, we investigate the bottlenecks in current MLLMs and synthesize training data to improve their abstract visual perception. First, we propose VisuRiddles, a benchmark for AVR, featuring tasks meticulously constructed to assess models' reasoning capacities across five core dimensions and two high-level reasoning categories. Second, we introduce the Perceptual Riddle Synthesizer (PRS), an automated framework for generating riddles with fine-grained perceptual descriptions. PRS not only generates valuable training data for abstract graphics but also provides fine-grained perceptual description, crucially allowing for supervision over intermediate reasoning stages and thereby improving both training efficacy and model interpretability. Our extensive experimental results on VisuRiddles empirically validate that fine-grained visual perception is the principal bottleneck and our synthesis framework markedly enhances the performance of contemporary MLLMs on these challenging tasks. Our code and dataset will be released at https://github.com/yh-hust/VisuRiddles

Yunming Liang, Zihao Chen, Chaofan Ding et al.

Currently, high-quality, synchronized audio is synthesized from video and optional text inputs using various multi-modal joint learning frameworks. However, the precise alignment between the visual and generated audio domains remains far from satisfactory. One key factor is the lack of sufficient temporal and semantic alignment annotations in open-source video-audio and text-audio benchmarks. Therefore, we propose a framework for audio generation from videos, leveraging the internal chain-of-thought (CoT) of a multi-modal large language model (MLLM) to enable step-by-step reasoning without requiring additional annotations. Additionally, a corresponding multi-modal reasoning dataset is constructed to facilitate the learning of initial reasoning in audio generation. In the experiments, we demonstrate the effectiveness of the proposed framework in reducing misalignment (voice-over) in generated audio and achieving competitive performance compared to various state-of-the-art models. The evaluation results show that the proposed method outperforms state-of-the-art approaches across multiple metrics. Specifically, the F DP aSST indicator is reduced by up to 10.07%, the F DP AN N s indicator by up to 11.62%, and the F DV GG indicator by up to 38.61%. Furthermore, the IS indicator improves by up to 4.95%, the IB-score indicator increases by up to 6.39%, and the DeSync indicator is reduced by up to 0.89%.

Xinhan Di, Zihao Chen, Yunming Liang et al.

Large-scale text-to-speech (TTS) models have made significant progress recently.However, they still fall short in the generation of Chinese dialectal speech. Toaddress this, we propose Bailing-TTS, a family of large-scale TTS models capable of generating high-quality Chinese dialectal speech. Bailing-TTS serves as a foundation model for Chinese dialectal speech generation. First, continual semi-supervised learning is proposed to facilitate the alignment of text tokens and speech tokens. Second, the Chinese dialectal representation learning is developed using a specific transformer architecture and multi-stage training processes. With the proposed design of novel network architecture and corresponding strategy, Bailing-TTS is able to generate Chinese dialectal speech from text effectively and efficiently. Experiments demonstrate that Bailing-TTS generates Chinese dialectal speech towards human-like spontaneous representation. Readers are encouraged to listen to demos at \url{https://c9412600.github.io/bltts_tech_report/index.html}.

Zihao Chen, Ji Zhuang, Jinyi Shen et al.

In this paper, we propose AnalogSeeker, an effort toward an open-source foundation language model for analog circuit design, with the aim of integrating domain knowledge and giving design assistance. To overcome the scarcity of data in this field, we employ a corpus collection strategy based on the domain knowledge framework of analog circuits. High-quality, accessible textbooks across relevant subfields are systematically curated and cleaned into a textual domain corpus. To address the complexity of knowledge of analog circuits, we introduce a granular domain knowledge distillation method. Raw, unlabeled domain corpus is decomposed into typical, granular learning nodes, where a multi-agent framework distills implicit knowledge embedded in unstructured text into question-answer data pairs with detailed reasoning processes, yielding a fine-grained, learnable dataset for fine-tuning. To address the unexplored challenges in training analog circuit foundation models, we explore and share our training methods through both theoretical analysis and experimental validation. We finally establish a fine-tuning-centric training paradigm, customizing and implementing a neighborhood self-constrained supervised fine-tuning algorithm. This approach enhances training outcomes by constraining the perturbation magnitude between the model's output distributions before and after training. In practice, we train the Qwen2.5-32B-Instruct model to obtain AnalogSeeker, which achieves 85.04% accuracy on AMSBench-TQA, the analog circuit knowledge evaluation benchmark, with a 15.67% point improvement over the original model and is competitive with mainstream commercial models. Furthermore, AnalogSeeker also shows effectiveness in the downstream operational amplifier design task. AnalogSeeker is open-sourced at https://huggingface.co/analogllm/analogseeker for research use.

Zihao Chen, Hsuanyu Wu, Chi-Hsi Kung et al.

Traffic Atomic Activity which describes traffic patterns for topological intersection dynamics is a crucial topic for the advancement of intelligent driving systems. However, existing atomic activity datasets are collected from an egocentric view, which cannot support the scenarios where traffic activities in an entire intersection are required. Moreover, existing datasets only provide video-level atomic activity annotations, which require exhausting efforts to manually trim the videos for recognition and limit their applications to untrimmed videos. To bridge this gap, we introduce the Aerial Traffic Atomic Activity Recognition and Segmentation (ATARS) dataset, the first aerial dataset designed for multi-label atomic activity analysis. We offer atomic activity labels for each frame, which accurately record the intervals for traffic activities. Moreover, we propose a novel task, Multi-label Temporal Atomic Activity Recognition, enabling the study of accurate temporal localization for atomic activity and easing the burden of manual video trimming for recognition. We conduct extensive experiments to evaluate existing state-of-the-art models on both atomic activity recognition and temporal atomic activity segmentation. The results highlight the unique challenges of our ATARS dataset, such as recognizing extremely small objects' activities. We further provide comprehensive discussion analyzing these challenges and offer valuable insights for future direction to improve recognizing atomic activity in aerial view. Our source code and dataset are available at https://github.com/magecliff96/ATARS/

Zihao Chen, Hisashi Handa, Miho Ohsaki et al.

Several backbone models pre-trained on general domain datasets can encode a sentence into a widely useful embedding. Such sentence embeddings can be further enhanced by domain adaptation that adapts a backbone model to a specific domain. However, domain adaptation for low-resource languages like Japanese is often difficult due to the scarcity of large-scale labeled datasets. To overcome this, this paper introduces SDJC (Self-supervised Domain adaptation for Japanese sentence embeddings with Contrastive learning) that utilizes a data generator to generate sentences, which have the same syntactic structure to a sentence in an unlabeled specific domain corpus but convey different semantic meanings. Generated sentences are then used to boost contrastive learning that adapts a backbone model to accurately discriminate sentences in the specific domain. In addition, the components of SDJC like a backbone model and a method to adapt it need to be carefully selected, but no benchmark dataset is available for Japanese. Thus, a comprehensive Japanese STS (Semantic Textual Similarity) benchmark dataset is constructed by combining datasets machine-translated from English with existing datasets. The experimental results validates the effectiveness of SDJC on two domain-specific downstream tasks as well as the usefulness of the constructed dataset. Datasets, codes and backbone models adapted by SDJC are available on our github repository https://github.com/ccilab-doshisha/SDJC.

Zihao Chen, Zhentao Lin, Bi Zeng et al.

To ensure the reliable operation of speech systems across diverse environments, noise addition methods have emerged as the standard solution.However, existing methods offer limited coverage of real-world scenes and depend on pre-existing noise libraries and scene metadata.This paper presents prompt-based Dynamic Generative Scene-based Noise Addition (DGSNA), a novel approach driven by generative language models that integrates Dynamic Generation of Scene-based Information (DGSI) with Scene-based Noise Addition for Speech (SNAS).The DGSI module, with a BET (Background, Examples, Task) prompt framework, dynamically generates logic-compliant scene-based information, including scene dimensions, sound sources, and microphone positions, thereby addressing the challenges of scene enumeration and detailed description.Complementing this, the SNAS module employs a Time-Frequency Diffusion-based (TFD) Text-to-Audio model to synthesize scene-specific noise. By integrating this noise with clean speech via Room Impulse Response (RIR) filters, the module streamlines the traditionally labor-intensive process of replicating diverse acoustic environments.Experimental results show that DGSNA significantly enhances the robustness of speech recognition and keyword spotting models, achieving relative improvements of up to 11.32\%. Furthermore, DGSNA is highly compatible with existing noise addition techniques. Our implementation and demonstrations are available at https://dgsna.github.io.

Jingze Song, Zihao Chen, Wenqing Chen et al.

Data selection is a key component of efficient instruction tuning for large language models, as recent work has shown that data quality often matters more than data quantity. Accordingly, prior studies have introduced various multi-dimensional heuristics to evaluate and filter instruction data. However, most existing methods rely on static task-agnostic and model-agnostic weighting schemes, which overlook the varying requirements of specific downstream tasks and the differing pre-existing capabilities of models. In this paper, we propose a framework for learning multi-indicator weights that jointly adapts data selection to both the downstream task and the specific model. Our method identifies optimal weight configurations without full-scale fine-tuning by utilizing in-context learning (ICL) signals on compact tiny-validation sets. These signals serve as efficient performance proxies that ensure high-fidelity evaluation at minimal computational cost. Experiments across multiple benchmarks and model families, including Mistral, Qwen, and Llama, show that the approach achieves performance comparable to or exceeding full-dataset tuning while using only 30\% of the training samples on GSM8K. Furthermore, our analysis reveals a trade-off between semantic diversity and logical complexity in reasoning tasks, highlighting the necessity of joint task-model adaptation.

Junjie Zheng, Chunbo Hao, Guobin Ma et al.

Singing Voice Synthesis (SVS) remains constrained in practical deployment due to its strong dependence on accurate phoneme-level alignment and manually annotated melody contours, requirements that are resource-intensive and hinder scalability. To overcome these limitations, we propose a melody-driven SVS framework capable of synthesizing arbitrary lyrics following any reference melody, without relying on phoneme-level alignment. Our method builds on a Diffusion Transformer (DiT) architecture, enhanced with a dedicated melody extraction module that derives melody representations directly from reference audio. To ensure robust melody encoding, we employ a teacher model to guide the optimization of the melody extractor, alongside an implicit alignment mechanism that enforces similarity distribution constraints for improved melodic stability and coherence. Additionally, we refine duration modeling using weakly annotated song data and introduce a Flow-GRPO reinforcement learning strategy with a multi-objective reward function to jointly enhance pronunciation clarity and melodic fidelity. Experiments show that our model achieves superior performance over existing approaches in both objective measures and subjective listening tests, especially in zero-shot and lyric adaptation settings, while maintaining high audio quality without manual annotation. This work offers a practical and scalable solution for advancing data-efficient singing voice synthesis. To support reproducibility, we release our inference code and model checkpoints.

Xifeng Yao, Chengyuan Ma, Dongyu Lang et al.

In recent months, substantial progress has been made in complex reasoning of Large Language Models, particularly through the application of test-time scaling. Notable examples include o1/o3/o4 series and DeepSeek-R1. When responding to a query, these models generate an extended reasoning trajectory, during which the model explores, reflects, backtracks, and self-verifies before arriving at a conclusion. However, fine-tuning models with such reasoning trajectories may not always be optimal. Our findings indicate that not all components within these reasoning trajectories contribute positively to the reasoning process; in fact, some components may affect the overall performance negatively. In this study, we divide a reasoning trajectory into individual subtrajectories and develop a "5+2" framework to: (1) systematically identify suboptimal subtrajectories within the reasoning trajectory based on five human-established criteria; (2) assess the independence of the suboptimal subtrajectories identified in (1) from the subsequent content, ensuring that their elimination does not compromise overall flow and coherence of the reasoning process. Additionally, a sampling algorithm, built upon the "5+2" framework, is employed to select data whose reasoning process is free from suboptimal subtrajectories to the highest degree. Experimental results demonstrate that our method can reduce the number of suboptimal subtrajectories by 25.9\% during the inference. Furthermore, our method achieves an average accuracy of 58.92\% on highly challenging math benchmarks with only two thirds of training data, surpassing the average accuracy of 58.06\% achieved with the entire data, and outperforming open-source datasets, when fine-tuning Qwen2.5-Math-7B. Finally, We validated our method under resource constraints and observed improved performance across various inference token limits.

Haomin Zhang, Kristin Qi, Shuxin Yang et al.

Generating high-quality and temporally synchronized audio from video content is essential for video editing and post-production tasks, enabling the creation of semantically aligned audio for silent videos. However, most existing approaches focus on short-form audio generation for video segments under 10 seconds or rely on noisy datasets for long-form video-to-audio zsynthesis. To address these limitations, we introduce LD-LAudio-V1, an extension of state-of-the-art video-to-audio models and it incorporates dual lightweight adapters to enable long-form audio generation. In addition, we release a clean and human-annotated video-to-audio dataset that contains pure sound effects without noise or artifacts. Our method significantly reduces splicing artifacts and temporal inconsistencies while maintaining computational efficiency. Compared to direct fine-tuning with short training videos, LD-LAudio-V1 achieves significant improvements across multiple metrics: $FD_{\text{passt}}$ 450.00 $\rightarrow$ 327.29 (+27.27%), $FD_{\text{panns}}$ 34.88 $\rightarrow$ 22.68 (+34.98%), $FD_{\text{vgg}}$ 3.75 $\rightarrow$ 1.28 (+65.87%), $KL_{\text{panns}}$ 2.49 $\rightarrow$ 2.07 (+16.87%), $KL_{\text{passt}}$ 1.78 $\rightarrow$ 1.53 (+14.04%), $IS_{\text{panns}}$ 4.17 $\rightarrow$ 4.30 (+3.12%), $IB_{\text{score}}$ 0.25 $\rightarrow$ 0.28 (+12.00%), $Energy\Delta10\text{ms}$ 0.3013 $\rightarrow$ 0.1349 (+55.23%), $Energy\Delta10\text{ms(vs.GT)}$ 0.0531 $\rightarrow$ 0.0288 (+45.76%), and $Sem.\,Rel.$ 2.73 $\rightarrow$ 3.28 (+20.15%). Our dataset aims to facilitate further research in long-form video-to-audio generation and is available at https://github.com/deepreasonings/long-form-video2audio.

Chaoyi Wang, Junjie Zheng, Zihao Chen et al.

Movie dubbing has advanced significantly, yet assessing the real-world effectiveness of these models remains challenging. A comprehensive evaluation benchmark is crucial for two key reasons: 1) Existing metrics fail to fully capture the complexities of dialogue, narration, monologue, and actor adaptability in movie dubbing. 2) A practical evaluation system should offer valuable insights to improve movie dubbing quality and advancement in film production. To this end, we introduce Talking Adaptive Dubbing Benchmarks (TA-Dubbing), designed to improve film production by adapting to dialogue, narration, monologue, and actors in movie dubbing. TA-Dubbing offers several key advantages: 1) Comprehensive Dimensions: TA-Dubbing covers a variety of dimensions of movie dubbing, incorporating metric evaluations for both movie understanding and speech generation. 2) Versatile Benchmarking: TA-Dubbing is designed to evaluate state-of-the-art movie dubbing models and advanced multi-modal large language models. 3) Full Open-Sourcing: We fully open-source TA-Dubbing at https://github.com/woka- 0a/DeepDubber- V1 including all video suits, evaluation methods, annotations. We also continuously integrate new movie dubbing models into the TA-Dubbing leaderboard at https://github.com/woka- 0a/DeepDubber-V1 to drive forward the field of movie dubbing.

Mostafa Elhoushi, Zihao Chen, Farhan Shafiq et al.

The high computation, memory, and power budgets of inferring convolutional neural networks (CNNs) are major bottlenecks of model deployment to edge computing platforms, e.g., mobile devices and IoT. Moreover, training CNNs is time and energy-intensive even on high-grade servers. Convolution layers and fully connected layers, because of their intense use of multiplications, are the dominant contributor to this computation budget. We propose to alleviate this problem by introducing two new operations: convolutional shifts and fully-connected shifts which replace multiplications with bitwise shift and sign flipping during both training and inference. During inference, both approaches require only 5 bits (or less) to represent the weights. This family of neural network architectures (that use convolutional shifts and fully connected shifts) is referred to as DeepShift models. We propose two methods to train DeepShift models: DeepShift-Q which trains regular weights constrained to powers of 2, and DeepShift-PS that trains the values of the shifts and sign flips directly. Very close accuracy, and in some cases higher accuracy, to baselines are achieved. Converting pre-trained 32-bit floating-point baseline models of ResNet18, ResNet50, VGG16, and GoogleNet to DeepShift and training them for 15 to 30 epochs, resulted in Top-1/Top-5 accuracies higher than that of the original model. Last but not least, we implemented the convolutional shifts and fully connected shift GPU kernels and showed a reduction in latency time of 25% when inferring ResNet18 compared to unoptimized multiplication-based GPU kernels. The code can be found at https://github.com/mostafaelhoushi/DeepShift.

Jiahui Chen, Weida Wang, Runhua Shi et al.

While diffusion model for audio-driven avatar video generation have achieved notable process in synthesizing long sequences with natural audio-visual synchronization and identity consistency, the generation of music-performance videos with camera motions remains largely unexplored. We present YingVideo-MV, the first cascaded framework for music-driven long-video generation. Our approach integrates audio semantic analysis, an interpretable shot planning module (MV-Director), temporal-aware diffusion Transformer architectures, and long-sequence consistency modeling to enable automatic synthesis of high-quality music performance videos from audio signals. We construct a large-scale Music-in-the-Wild Dataset by collecting web data to support the achievement of diverse, high-quality results. Observing that existing long-video generation methods lack explicit camera motion control, we introduce a camera adapter module that embeds camera poses into latent noise. To enhance continulity between clips during long-sequence inference, we further propose a time-aware dynamic window range strategy that adaptively adjust denoising ranges based on audio embedding. Comprehensive benchmark tests demonstrate that YingVideo-MV achieves outstanding performance in generating coherent and expressive music videos, and enables precise music-motion-camera synchronization. More videos are available in our project page: https://giantailab.github.io/YingVideo-MV/ .

Zihao Chen, Haomin Zhang, Xinhan Di et al.

Generating sound effects for product-level videos, where only a small amount of labeled data is available for diverse scenes, requires the production of high-quality sounds in few-shot settings. To tackle the challenge of limited labeled data in real-world scenes, we introduce YingSound, a foundation model designed for video-guided sound generation that supports high-quality audio generation in few-shot settings. Specifically, YingSound consists of two major modules. The first module uses a conditional flow matching transformer to achieve effective semantic alignment in sound generation across audio and visual modalities. This module aims to build a learnable audio-visual aggregator (AVA) that integrates high-resolution visual features with corresponding audio features at multiple stages. The second module is developed with a proposed multi-modal visual-audio chain-of-thought (CoT) approach to generate finer sound effects in few-shot settings. Finally, an industry-standard video-to-audio (V2A) dataset that encompasses various real-world scenarios is presented. We show that YingSound effectively generates high-quality synchronized sounds across diverse conditional inputs through automated evaluations and human studies. Project Page: \url{https://giantailab.github.io/yingsound/}

Haomin Zhang, Chang Liu, Junjie Zheng et al.

Currently, high-quality, synchronized audio is synthesized using various multi-modal joint learning frameworks, leveraging video and optional text inputs. In the video-to-audio benchmarks, video-to-audio quality, semantic alignment, and audio-visual synchronization are effectively achieved. However, in real-world scenarios, speech and audio often coexist in videos simultaneously, and the end-to-end generation of synchronous speech and audio given video and text conditions are not well studied. Therefore, we propose an end-to-end multi-modal generation framework that simultaneously produces speech and audio based on video and text conditions. Furthermore, the advantages of video-to-audio (V2A) models for generating speech from videos remain unclear. The proposed framework, DeepAudio, consists of a video-to-audio (V2A) module, a text-to-speech (TTS) module, and a dynamic mixture of modality fusion (MoF) module. In the evaluation, the proposed end-to-end framework achieves state-of-the-art performance on the video-audio benchmark, video-speech benchmark, and text-speech benchmark. In detail, our framework achieves comparable results in the comparison with state-of-the-art models for the video-audio and text-speech benchmarks, and surpassing state-of-the-art models in the video-speech benchmark, with WER 16.57% to 3.15% (+80.99%), SPK-SIM 78.30% to 89.38% (+14.15%), EMO-SIM 66.24% to 75.56% (+14.07%), MCD 8.59 to 7.98 (+7.10%), MCD SL 11.05 to 9.40 (+14.93%) across a variety of dubbing settings.

Junjie Zheng, Zihao Chen, Chaofan Ding et al.

Current movie dubbing technology can generate the desired voice from a given speech prompt, ensuring good synchronization between speech and visuals while accurately conveying the intended emotions. However, in movie dubbing, key aspects such as adapting to different dubbing styles, handling dialogue, narration, and monologue effectively, and understanding subtle details like the age and gender of speakers, have not been well studied. To address this challenge, we propose a framework of multi-modal large language model. First, it utilizes multimodal Chain-of-Thought (CoT) reasoning methods on visual inputs to understand dubbing styles and fine-grained attributes. Second, it generates high-quality dubbing through large speech generation models, guided by multimodal conditions. Additionally, we have developed a movie dubbing dataset with CoT annotations. The evaluation results demonstrate a performance improvement over state-of-the-art methods across multiple datasets. In particular, for the evaluation metrics, the SPK-SIM and EMO-SIM increases from 82.48% to 89.74%, 66.24% to 78.88% for dubbing setting 2.0 on V2C Animation dataset, LSE-D and MCD-SL decreases from 14.79 to 14.63, 5.24 to 4.74 for dubbing setting 2.0 on Grid dataset, SPK-SIM increases from 64.03 to 83.42 and WER decreases from 52.69% to 23.20% for initial reasoning setting on proposed CoT-Movie-Dubbing dataset in the comparison with the state-of-the art models.

Zihao Chen, Chi-Heng Lin, Ran Liu et al.

Despite the success of contrastive learning (CL) in vision and language, its theoretical foundations and mechanisms for building representations remain poorly understood. In this work, we build connections between noise contrastive estimation losses widely used in CL and distribution alignment with entropic optimal transport (OT). This connection allows us to develop a family of different losses and multistep iterative variants for existing CL methods. Intuitively, by using more information from the distribution of latents, our approach allows a more distribution-aware manipulation of the relationships within augmented sample sets. We provide theoretical insights and experimental evidence demonstrating the benefits of our approach for {\em generalized contrastive alignment}. Through this framework, it is possible to leverage tools in OT to build unbalanced losses to handle noisy views and customize the representation space by changing the constraints on alignment. By reframing contrastive learning as an alignment problem and leveraging existing optimization tools for OT, our work provides new insights and connections between different self-supervised learning models in addition to new tools that can be more easily adapted to incorporate domain knowledge into learning.

Zihao Chen, Faiek Ahsan, Johannes Leugering et al.

Neuromorphic or neurally-inspired optimizers rely on local but parallel parameter updates to solve problems that range from quadratic programming to Ising machines. An ideal realization of such an optimizer not only uses a compute-in-memory (CIM) paradigm to address the so-called memory-wall (i.e. energy dissipated due to repeated memory read access), but also uses a learning-in-memory (LIM) paradigm to address the energy bottlenecks due to repeated memory writes at the precision required for optimization (the update-wall), and to address the energy bottleneck due to the repeated transfer of information between short-term and long-term memories (the consolidation-wall). In this paper, we derive theoretical estimates for the energy-to-solution metric that can be achieved by this ideal neuromorphic optimizer which is realized by modulating the energy-barrier of the physical memories such that the dynamics of memory updates and memory consolidation matches the optimization or the annealing dynamics. The analysis presented in this paper captures the out-of-equilibrium thermodynamics of learning and the resulting energy-efficiency estimates are model-agnostic which only depend on the number of model-update operations (OPS), the model-size in terms of number of parameters, the speed of convergence, and the precision of the solution. To show the practical applicability of our results, we apply our analysis for estimating the lower-bound on the energy-to-solution metrics for large-scale AI workloads.

Junjie Zheng, Gongyu Chen, Chaofan Ding et al.

In real-world singing voice conversion (SVC) applications, environmental noise and the demand for expressive output pose significant challenges. Conventional methods, however, are typically designed without accounting for real deployment scenarios, as both training and inference usually rely on clean data. This mismatch hinders practical use, given the inevitable presence of diverse noise sources and artifacts from music separation. To tackle these issues, we propose R2-SVC, a robust and expressive SVC framework. First, we introduce simulation-based robustness enhancement through random fundamental frequency ($F_0$) perturbations and music separation artifact simulations (e.g., reverberation, echo), substantially improving performance under noisy conditions. Second, we enrich speaker representation using domain-specific singing data: alongside clean vocals, we incorporate DNSMOS-filtered separated vocals and public singing corpora, enabling the model to preserve speaker timbre while capturing singing style nuances. Third, we integrate the Neural Source-Filter (NSF) model to explicitly represent harmonic and noise components, enhancing the naturalness and controllability of converted singing. R2-SVC achieves state-of-the-art results on multiple SVC benchmarks under both clean and noisy conditions.

Zihao Chen, Yi Zhou, Xudong Jiang et al.

Unpaired image-to-image translation has emerged as a crucial technique in medical imaging, enabling cross-modality synthesis, domain adaptation, and data augmentation without costly paired datasets. Yet, existing approaches often distort fine curvilinear structures, such as microvasculature, undermining both diagnostic reliability and quantitative analysis. This limitation is consequential in ophthalmic and vascular imaging, where subtle morphological changes carry significant clinical meaning. We propose Curvilinear Structure-preserving Translation (CST), a general framework that explicitly preserves fine curvilinear structures during unpaired translation by integrating structure consistency into the training. Specifically, CST augments baseline models with a curvilinear extraction module for topological supervision. It can be seamlessly incorporated into existing methods. We integrate it into CycleGAN and UNSB as two representative backbones. Comprehensive evaluation across three imaging modalities: optical coherence tomography angiography, color fundus and X-ray coronary angiography demonstrates that CST improves translation fidelity and achieves state-of-the-art performance. By reinforcing geometric integrity in learned mappings, CST establishes a principled pathway toward curvilinear structure-aware cross-domain translation in medical imaging.

Ziqi Chen, Gongyu Chen, Yihua Wang et al.

Dialect speech embodies rich cultural and linguistic diversity, yet building text-to-speech (TTS) systems for dialects remains challenging due to scarce data, inconsistent orthographies, and complex phonetic variation. To address these issues, we present DiaMoE-TTS, a unified IPA-based framework that standardizes phonetic representations and resolves grapheme-to-phoneme ambiguities. Built upon the F5-TTS architecture, the system introduces a dialect-aware Mixture-of-Experts (MoE) to model phonological differences and employs parameter-efficient adaptation with Low-Rank Adaptors (LoRA) and Conditioning Adapters for rapid transfer to new dialects. Unlike approaches dependent on large-scale or proprietary resources, DiaMoE-TTS enables scalable, open-data-driven synthesis. Experiments demonstrate natural and expressive speech generation, achieving zero-shot performance on unseen dialects and specialized domains such as Peking Opera with only a few hours of data.

Junjie Zheng, Zihao Chen, Chaofan Ding et al.

Current movie dubbing technology can produce the desired speech using a reference voice and input video, maintaining perfect synchronization with the visuals while effectively conveying the intended emotions. However, crucial aspects of movie dubbing, including adaptation to various dubbing styles, effective handling of dialogue, narration, and monologues, as well as consideration of subtle details such as speaker age and gender, remain insufficiently explored. To tackle these challenges, we introduce a multi-modal generative framework. First, it utilizes a multi-modal large vision-language model (VLM) to analyze visual inputs, enabling the recognition of dubbing types and fine-grained attributes. Second, it produces high-quality dubbing using large speech generation models, guided by multi-modal inputs. Additionally, a movie dubbing dataset with annotations for dubbing types and subtle details is constructed to enhance movie understanding and improve dubbing quality for the proposed multi-modal framework. Experimental results across multiple benchmark datasets show superior performance compared to state-of-the-art (SOTA) methods. In details, the LSE-D, SPK-SIM, EMO-SIM, and MCD exhibit improvements of up to 1.09%, 8.80%, 19.08%, and 18.74%, respectively.

Zihao Chen, Yu Xiang, Wenyong Wang

Despite the success in learning semantically meaningful, unsupervised disentangled representations, variational autoencoders (VAEs) and their variants face a fundamental theoretical challenge: substantial evidence indicates that unsupervised disentanglement is unattainable without implicit inductive bias, yet such bias remains elusive. In this work, we focus on exploring the implicit inductive bias that drive disentanglement in VAEs with factorization priors. By analyzing the total correlation in \b{eta}-TCVAE, we uncover a crucial implicit inductive bias called disentangling granularity, which leads to the discovery of an interesting "V"-shaped optimal Evidence Lower Bound (ELBO) trajectory within the parameter space. This finding is validated through over 100K experiments using factorized VAEs and our newly proposed model, \b{eta}-STCVAE. Notably, experimental results reveal that conventional factorized VAEs, constrained by fixed disentangling granularity, inherently tend to disentangle low-complexity feature. Whereas, appropriately tuning disentangling granularity, as enabled by \b{eta}-STCVAE, broadens the range of disentangled representations, allowing for the disentanglement of high-complexity features. Our findings unveil that disentangling granularity as an implicit inductive bias in factorized VAEs influence both disentanglement performance and the inference of the ELBO, offering fresh insights into the interpretability and inherent biases of VAEs.

Zihao Chen, Wenyong Wang, Jiachen Yang et al.

Geometric representation learning in preserving the intrinsic geometric and topological properties for discrete non-Euclidean data is crucial in scientific applications. Previous research generally mapped non-Euclidean discrete data into Euclidean space during representation learning, which may lead to the loss of some critical geometric information. In this paper, we propose a novel Isometric Immersion Kernel Learning (IIKL) method to build Riemannian manifold and isometrically induce Riemannian metric from discrete non-Euclidean data. We prove that Isometric immersion is equivalent to the kernel function in the tangent bundle on the manifold, which explicitly guarantees the invariance of the inner product between vectors in the arbitrary tangent space throughout the learning process, thus maintaining the geometric structure of the original data. Moreover, a novel parameterized learning model based on IIKL is introduced, and an alternating training method for this model is derived using Maximum Likelihood Estimation (MLE), ensuring efficient convergence. Experimental results proved that using the learned Riemannian manifold and its metric, our model preserved the intrinsic geometric representation of data in both 3D and high-dimensional datasets successfully, and significantly improved the accuracy of downstream tasks, such as data reconstruction and classification. It is showed that our method could reduce the inner product invariant loss by more than 90% compared to state-of-the-art (SOTA) methods, also achieved an average 40% improvement in downstream reconstruction accuracy and a 90% reduction in error for geometric metrics involving isometric and conformal.

Haomin Zhang, Sizhe Shan, Haoyu Wang et al.

Creating high-quality sound effects from videos and text prompts requires precise alignment between visual and audio domains, both semantically and temporally, along with step-by-step guidance for professional audio generation. However, current state-of-the-art video-guided audio generation models often fall short of producing high-quality audio for both general and specialized use cases. To address this challenge, we introduce a multi-stage, multi-modal, end-to-end generative framework with Chain-of-Thought-like (CoT-like) guidance learning, termed Chain-of-Perform (CoP). First, we employ a transformer-based network architecture designed to achieve CoP guidance, enabling the generation of both general and professional audio. Second, we implement a multi-stage training framework that follows step-by-step guidance to ensure the generation of high-quality sound effects. Third, we develop a CoP multi-modal dataset, guided by video, to support step-by-step sound effects generation. Evaluation results highlight the advantages of the proposed multi-stage CoP generative framework compared to the state-of-the-art models on a variety of datasets, with FAD 0.79 to 0.74 (+6.33%), CLIP 16.12 to 17.70 (+9.80%) on VGGSound, SI-SDR 1.98dB to 3.35dB (+69.19%), MOS 2.94 to 3.49(+18.71%) on PianoYT-2h, and SI-SDR 2.22dB to 3.21dB (+44.59%), MOS 3.07 to 3.42 (+11.40%) on Piano-10h.

Faiek Ahsan, Saptarshi Maiti, Zihao Chen et al.

We report a higher-order neuromorphic Ising machine that exhibits superior scalability compared to architectures based on quadratization, while also achieving state-of-the-art quality and reliability in solutions with competitive time-to-solution metrics. At the core of the proposed machine is an asynchronous autoencoder architecture that captures higher-order interactions by directly manipulating Ising clauses instead of Ising spins, thereby maintaining resource complexity independent of interaction order. Asymptotic convergence to the Ising ground state is ensured by sampling the autoencoder latent space defined by the spins, based on the annealing dynamics of the Fowler-Nordheim quantum mechanical tunneling. To demonstrate the advantages of the proposed higher-order neuromorphic Ising machine, we systematically solved benchmark combinatorial optimization problems such as MAX-CUT and MAX-SAT, comparing the results to those obtained using a second-order Ising machine employing the same annealing process. Our findings indicate that the proposed architecture consistently provides higher quality solutions in shorter time frames compared to the second-order model across multiple runs. Additionally, we show that the techniques based on the sparsity of the interconnection matrix, such as graph coloring, can be effectively applied to higher-order neuromorphic Ising machines, enhancing the solution quality and the time-to-solution. The time-to-solution can be further improved through hardware co-design, as demonstrated in this paper using a field-programmable gate array (FPGA). The results presented in this paper provide further evidence that autoencoders and Fowler-Nordheim annealers are sufficient to achieve reliability and scaling of any-order neuromorphic Ising machines.

Hongxi Pu, Futian Jiang, Zihao Chen et al.

Music composition has long been recognized as a significant art form. However, existing digital audio workstations and music production software often present high entry barriers for users lacking formal musical training. To address this, we introduce ComposeOn, a music theory-based tool designed for users with limited musical knowledge. ComposeOn enables users to easily extend their melodic ideas into complete compositions and offers simple editing features. By integrating music theory, it explains music creation at beginner, intermediate, and advanced levels. Our user study (N=10) compared ComposeOn with the baseline method, Suno AI, demonstrating that ComposeOn provides a more accessible and enjoyable composing and learning experience for individuals with limited musical skills. ComposeOn bridges the gap between theory and practice, offering an innovative solution as both a composition aid and music education platform. The study also explores the differences between theory-based music creation and generative music, highlighting the former's advantages in personal expression and learning.

Gongyu Chen, Haomin Zhang, Chaofan Ding et al.

One fascinating aspect of pre-trained Audio-Language Models (ALMs) learning is their impressive zero-shot generalization capability and test-time adaptation (TTA) methods aiming to improve domain performance without annotations. However, previous test time adaptation (TTA) methods for ALMs in zero-shot classification tend to be stuck in incorrect model predictions. In order to further boost the performance, we propose multiple guidance on prompt learning without annotated labels. First, guidance of consistency on both context tokens and domain tokens of ALMs is set. Second, guidance of both consistency across multiple augmented views of each single test sample and contrastive learning across different test samples is set. Third, we propose a corresponding end-end learning framework for the proposed test-time adaptation method without annotated labels. We extensively evaluate our approach on 12 downstream tasks across domains, our proposed adaptation method leads to 4.41% (max 7.50%) average zero-shot performance improvement in comparison with the state-of-the-art models.

Wenwei Xie, Jie Yin, Zihao Chen

To address the issues of insufficient robustness, unstable features, and data noise interference in existing network attack detection and identification models, this paper proposes an attack traffic detection and identification method based on temporal spectrum. First, traffic data is segmented by a sliding window to construct a feature sequence and a corresponding label sequence for network traffic. Next, the proposed spectral label generation methods, SSPE and COAP, are applied to transform the label sequence into spectral labels and the feature sequence into temporal features. Spectral labels and temporal features are used to capture and represent behavioral patterns of attacks. Finally, the constructed temporal features and spectral labels are used to train models, which subsequently detects and identifies network attack behaviors. Experimental results demonstrate that compared to traditional methods, models trained with the SSPE or COAP method improve identification accuracy by 10%, and exhibit strong robustness, particularly in noisy environments.

Xin Chen, Yawen Duan, Zewei Chen et al.

Neural Architecture Search (NAS) achieved many breakthroughs in recent years. In spite of its remarkable progress, many algorithms are restricted to particular search spaces. They also lack efficient mechanisms to reuse knowledge when confronting multiple tasks. These challenges preclude their applicability, and motivate our proposal of CATCH, a novel Context-bAsed meTa reinforcement learning (RL) algorithm for transferrable arChitecture searcH. The combination of meta-learning and RL allows CATCH to efficiently adapt to new tasks while being agnostic to search spaces. CATCH utilizes a probabilistic encoder to encode task properties into latent context variables, which then guide CATCH's controller to quickly "catch" top-performing networks. The contexts also assist a network evaluator in filtering inferior candidates and speed up learning. Extensive experiments demonstrate CATCH's universality and search efficiency over many other widely-recognized algorithms. It is also capable of handling cross-domain architecture search as competitive networks on ImageNet, COCO, and Cityscapes are identified. This is the first work to our knowledge that proposes an efficient transferrable NAS solution while maintaining robustness across various settings.