arXiv subjects

Weikang Wang

Publications and source records attributed to Weikang Wang.

At least 19 recordsLinked to original sources

arXiv2026

Unsupervised Pixel-Level Semantic Left-Right Understanding of In-the-Wild Images

While various works address reflective symmetry understanding in 3D data and images, pixel-level semantic left-right prediction of in-the-wild images remains challenging, due to certain difficulties including the lack of 3D information, occlusion, object pose variation, partiality, etc. In this work, we propose an unsupervised learning framework to tackle this challenge. Leveraging recent advances in vertex-wise semantic left-right understanding of 3D data, our unsupervised learning method jointly utilises 3D shape and image datasets to infer pixel-wise semantic left-right predictions in single-view images. In particular, we show that a medium-scale 3D shape dataset comprising mainly of human- and quadruped animal-like shapes, combined with diverse in-the-wild image data, are sufficient to achieve high-quality semantic left-right prediction in images, even for entirely unseen 3D object categories, such as cars or trains. Overall, our approach achieves superior performance in dense pixel-wise semantic left-right predictions on both rendered and in-the-wild image datasets when compared to existing state-of-the-art methods.

cs.CV↗

arXiv2026

Efficient coding under constraint drives neural systems towards criticality and sloppiness

It is widely accepted that the brain operates near a critical state, characterized by neural avalanches that follow power-law distributions. However, the functional rationale for why neural systems attain criticality remains unclear. Here, we present a theoretical framework that links efficient coding to criticality in neural populations. Using a Gaussian population coding model, we demonstrate that maximizing Fisher information under resource constraints naturally leads to the emergence of soft modes and diverging correlation lengths, which are hallmarks of criticality. By introducing spatial structure, we unify two distinct perspectives of criticality: statistical criticality with diverging correlation lengths and dynamical criticality with critical slowing down as well as bifurcation. Furthermore, this framework provides a natural explanation for the sloppiness observed in neural systems. Numerical simulations confirm that optimization results in power-law response, providing a mechanistic link between efficient coding, sloppiness and the critical brain hypothesis.

q-bio.NC↗

arXiv2026

Response time of lateral predictive coding and benefits of modular structures

Lateral predictive coding (LPC) is a simple theoretical framework to appreciate feature detection in biological neural circuits. Recent theoretical work [Huang et al., Phys.Rev.E 112, 034304 (2025)] has successfully constructed optimal LPC networks capable of extracting non-Gaussian hidden input features by imposing the tradeoff between energetic cost and information robustness, but the resulting dynamical systems of recurrent interactions can be very slow in responding to external inputs. We investigate response-time reduction in the present paper. We find that the characteristic response time of the LPC system can be minimized to closely approaching the lower-bound value without compromising the mean predictive error (energetic cost) and the information robustness of signal transmission. We further demonstrate that optimal LPC networks taking a modular structural organization with extensively reduced number of lateral interactions are equally excellent as all-to-all completely connected networks, in terms of feature detection performance, response time, energetic cost and information robustness.

q-bio.NC↗

arXiv2026

SPAGBias: Uncovering and Tracing Structured Spatial Gender Bias in Large Language Models

Large language models (LLMs) are being increasingly used in urban planning, but since gendered space theory highlights how gender hierarchies are embedded in spatial organization, there is concern that LLMs may reproduce or amplify such biases. We introduce SPAGBias - the first systematic framework to evaluate spatial gender bias in LLMs. It combines a taxonomy of 62 urban micro-spaces, a prompt library, and three diagnostic layers: explicit (forced-choice resampling), probabilistic (token-level asymmetry), and constructional (semantic and narrative role analysis). Testing six representative models, we identify structured gender-space associations that go beyond the public-private divide, forming nuanced micro-level mappings. Story generation reveals how emotion, wording, and social roles jointly shape "spatial gender narratives". We also examine how prompt design, temperature, and model scale influence bias expression. Tracing experiments indicate that these patterns are embedded and reinforced across the model pipeline (pre-training, instruction tuning, and reward modeling), with model associations found to substantially exceed real-world distributions. Downstream experiments further reveal that such biases produce concrete failures in both normative and descriptive application settings. This work connects sociological theory with computational analysis, extending bias research into the spatial domain and uncovering how LLMs encode social gender cognition through language.

cs.CL↗

arXiv2026

Symmetry Informative and Agnostic Feature Disentanglement for 3D Shapes

Shape descriptors, i.e., per-vertex features of 3D meshes or point clouds, are fundamental to shape analysis. Historically, various handcrafted geometry-aware descriptors and feature refinement techniques have been proposed. Recently, several studies have initiated a new research direction by leveraging features from image foundation models to create semantics-aware descriptors, demonstrating advantages across tasks like shape matching, editing, and segmentation. Symmetry, another key concept in shape analysis, has also attracted increasing attention. Consequently, constructing symmetry-aware shape descriptors is a natural progression. Although the recent method $\chi$ (Wang et al., 2025) successfully extracted symmetry-informative features from semantic-aware descriptors, its features are only one-dimensional, neglecting other valuable semantic information. Furthermore, the extracted symmetry-informative feature is usually noisy and yields small misclassified patches. To address these gaps, we propose a feature disentanglement approach which is simultaneously symmetry informative and symmetry agnostic. Further, we propose a feature refinement technique to improve the robustness of predicted symmetry informative features. Extensive experiments, including intrinsic symmetry detection, left/right classification, and shape matching, demonstrate the effectiveness of our proposed framework compared to various state-of-the-art methods, both qualitatively and quantitatively.

cs.CV↗

arXiv2025

Beyond Complete Shapes: A Benchmark for Quantitative Evaluation of 3D Shape Surface Matching Algorithms

Finding correspondences between 3D deformable shapes is an important and long-standing problem in geometry processing, computer vision, graphics, and beyond. While various shape matching datasets exist, they are mostly static or limited in size, restricting their adaptation to different problem settings, including both full and partial shape matching. In particular the existing partial shape matching datasets are small (fewer than 100 shapes) and thus unsuitable for data-hungry machine learning approaches. Moreover, the type of partiality present in existing datasets is often artificial and far from realistic. To address these limitations, we introduce a generic and flexible framework for the procedural generation of challenging full and partial shape matching datasets. Our framework allows the propagation of custom annotations across shapes, making it useful for various applications. By utilising our framework and manually creating cross-dataset correspondences between seven existing (complete geometry) shape matching datasets, we propose a new large benchmark BeCoS with a total of 2543 shapes. Based on this, we offer several challenging benchmark settings, covering both full and partial matching, for which we evaluate respective state-of-the-art methods as baselines.

cs.CV↗

arXiv2025

Discontinuous phase transition of feature detection in lateral predictive coding

The brain may adopt the strategy of lateral predictive coding (LPC) to construct optimal internal representations for salient features in input sensory signals, reducing the energetic cost of information transmission. Here we first consider the task of detecting one non-Gaussian signal by LPC from Gaussian background signals of the same magnitude, which is intractable by principal component decomposition. We study the emergence of feature detection function from the perspective of tradeoff between energetic cost $E$ and information robustness, and implement this tradeoff by a thermodynamic free energy. We define $E$ as the mean $L_1$-norm of the internal state vectors, and quantify the level of information robustness by an entropy measure $S$. There are at least three types of optimal LPC matrices, one type with very weak synaptic weights and $S \approx 0$, and two functional types either with low energy $E$ or with high entropy $S$ in which one single unit selectively responds to the non-Gaussian input feature. Energy--information tradeoff induce two discontinuous phase transitions between these three types of optimal LPC networks. We then extend the discussion to detecting and distinguishing between two non-Gaussian input features and observe similar discontinuous phase transitions.

q-bio.NC↗

arXiv2025

Symmetry Understanding of 3D Shapes via Chirality Disentanglement

Chirality information (i.e. information that allows distinguishing left from right) is ubiquitous for various data modes in computer vision, including images, videos, point clouds, and meshes. While chirality has been extensively studied in the image domain, its exploration in shape analysis (such as point clouds and meshes) remains underdeveloped. Although many shape vertex descriptors have shown appealing properties (e.g. robustness to rigid-body transformations), they are often not able to disambiguate between left and right symmetric parts. Considering the ubiquity of chirality information in different shape analysis problems and the lack of chirality-aware features within current shape descriptors, developing a chirality feature extractor becomes necessary and urgent. Based on the recent Diff3F framework, we propose an unsupervised chirality feature extraction pipeline to decorate shape vertices with chirality-aware information, extracted from 2D foundation models. We evaluated the extracted chirality features through quantitative and qualitative experiments across diverse datasets. Results from downstream tasks including left-right disentanglement, shape matching, and part segmentation demonstrate their effectiveness and practical utility. Project page: https://wei-kang-wang.github.io/chirality/

cs.CV↗

arXiv2025

Hunyuan-TurboS: Advancing Large Language Models through Mamba-Transformer Synergy and Adaptive Chain-of-Thought

As Large Language Models (LLMs) rapidly advance, we introduce Hunyuan-TurboS, a novel large hybrid Transformer-Mamba Mixture of Experts (MoE) model. It synergistically combines Mamba's long-sequence processing efficiency with Transformer's superior contextual understanding. Hunyuan-TurboS features an adaptive long-short chain-of-thought (CoT) mechanism, dynamically switching between rapid responses for simple queries and deep "thinking" modes for complex problems, optimizing computational resources. Architecturally, this 56B activated (560B total) parameter model employs 128 layers (Mamba2, Attention, FFN) with an innovative AMF/MF block pattern. Faster Mamba2 ensures linear complexity, Grouped-Query Attention minimizes KV cache, and FFNs use an MoE structure. Pre-trained on 16T high-quality tokens, it supports a 256K context length and is the first industry-deployed large-scale Mamba model. Our comprehensive post-training strategy enhances capabilities via Supervised Fine-Tuning (3M instructions), a novel Adaptive Long-short CoT Fusion method, Multi-round Deliberation Learning for iterative improvement, and a two-stage Large-scale Reinforcement Learning process targeting STEM and general instruction-following. Evaluations show strong performance: overall top 7 rank on LMSYS Chatbot Arena with a score of 1356, outperforming leading models like Gemini-2.0-Flash-001 (1352) and o4-mini-2025-04-16 (1345). TurboS also achieves an average of 77.9% across 23 automated benchmarks. Hunyuan-TurboS balances high performance and efficiency, offering substantial capabilities at lower inference costs than many reasoning models, establishing a new paradigm for efficient large-scale pre-trained models.

cs.CL↗

arXiv2025

HunyuanProver: A Scalable Data Synthesis Framework and Guided Tree Search for Automated Theorem Proving

We introduce HunyuanProver, an language model finetuned from the Hunyuan 7B for interactive automatic theorem proving with LEAN4. To alleviate the data sparsity issue, we design a scalable framework to iterative synthesize data with low cost. Besides, guided tree search algorithms are designed to enable effective ``system 2 thinking`` of the prover. HunyuanProver achieves state-of-the-art (SOTA) performances on major benchmarks. Specifically, it achieves a pass of 68.4% on the miniF2F-test compared to 65.9%, the current SOTA results. It proves 4 IMO statements (imo_1960_p2, imo_1962_p2}, imo_1964_p2 and imo_1983_p6) in miniF2F-test. To benefit the community, we will open-source a dataset of 30k synthesized instances, where each instance contains the original question in natural language, the converted statement by autoformalization, and the proof by HunyuanProver.

cs.AI↗

arXiv2024

Taking a Deep Breath: Enhancing Language Modeling of Large Language Models with Sentinel Tokens

Large language models (LLMs) have shown promising efficacy across various tasks, becoming powerful tools in numerous aspects of human life. However, Transformer-based LLMs suffer a performance degradation when modeling long-term contexts due to they discard some information to reduce computational overhead. In this work, we propose a simple yet effective method to enable LLMs to take a deep breath, encouraging them to summarize information contained within discrete text chunks. Specifically, we segment the text into multiple chunks and insert special token at the end of each chunk. We then modify the attention mask to integrate the chunk's information into the corresponding token. This facilitates LLMs to interpret information not only from historical individual tokens but also from the token, aggregating the chunk's semantic information. Experiments on language modeling and out-of-domain downstream tasks validate the superiority of our approach.

cs.CL↗

arXiv2024

Quite Good, but Not Enough: Nationality Bias in Large Language Models -- A Case Study of ChatGPT

While nationality is a pivotal demographic element that enhances the performance of language models, it has received far less scrutiny regarding inherent biases. This study investigates nationality bias in ChatGPT (GPT-3.5), a large language model (LLM) designed for text generation. The research covers 195 countries, 4 temperature settings, and 3 distinct prompt types, generating 4,680 discourses about nationality descriptions in Chinese and English. Automated metrics were used to analyze the nationality bias, and expert annotators alongside ChatGPT itself evaluated the perceived bias. The results show that ChatGPT's generated discourses are predominantly positive, especially compared to its predecessor, GPT-2. However, when prompted with negative inclinations, it occasionally produces negative content. Despite ChatGPT considering its generated text as neutral, it shows consistent self-awareness about nationality bias when subjected to the same pair-wise comparison annotation framework used by human annotators. In conclusion, while ChatGPT's generated texts seem friendly and positive, they reflect the inherent nationality biases in the real world. This bias may vary across different language versions of ChatGPT, indicating diverse cultural perspectives. The study highlights the subtle and pervasive nature of biases within LLMs, emphasizing the need for further scrutiny.

cs.CL↗

arXiv2023

Multilingual Sentence Transformer as A Multilingual Word Aligner

Multilingual pretrained language models (mPLMs) have shown their effectiveness in multilingual word alignment induction. However, these methods usually start from mBERT or XLM-R. In this paper, we investigate whether multilingual sentence Transformer LaBSE is a strong multilingual word aligner. This idea is non-trivial as LaBSE is trained to learn language-agnostic sentence-level embeddings, while the alignment extraction task requires the more fine-grained word-level embeddings to be language-agnostic. We demonstrate that the vanilla LaBSE outperforms other mPLMs currently used in the alignment task, and then propose to finetune LaBSE on parallel corpus for further improvement. Experiment results on seven language pairs show that our best aligner outperforms previous state-of-the-art models of all varieties. In addition, our aligner supports different language pairs in a single model, and even achieves new state-of-the-art on zero-shot language pairs that does not appear in the finetuning process.

cs.CL↗

arXiv2022

DS-Net: Dynamic Spatiotemporal Network for Video Salient Object Detection

As moving objects always draw more attention of human eyes, the temporal motive information is always exploited complementarily with spatial information to detect salient objects in videos. Although efficient tools such as optical flow have been proposed to extract temporal motive information, it often encounters difficulties when used for saliency detection due to the movement of camera or the partial movement of salient objects. In this paper, we investigate the complimentary roles of spatial and temporal information and propose a novel dynamic spatiotemporal network (DS-Net) for more effective fusion of spatiotemporal information. We construct a symmetric two-bypass network to explicitly extract spatial and temporal features. A dynamic weight generator (DWG) is designed to automatically learn the reliability of corresponding saliency branch. And a top-down cross attentive aggregation (CAA) procedure is designed so as to facilitate dynamic complementary aggregation of spatiotemporal features. Finally, the features are modified by spatial attention with the guidance of coarse saliency map and then go through decoder part for final saliency map. Experimental results on five benchmarks VOS, DAVIS, FBMS, SegTrack-v2, and ViSal demonstrate that the proposed method achieves superior performance than state-of-the-art algorithms. The source code is available at https://github.com/TJUMMG/DS-Net.

cs.CV↗

arXiv2022

Build Smart Grids on Artificial Intelligence -- A Real-world Example

Power grid data are going big with the deployment of various sensors. The big data in power grids creates huge opportunities for applying artificial intelligence technologies to improve resilience and reliability. This paper introduces multiple real-world applications based on artificial intelligence to improve power grid situational awareness and resilience. These applications include event identification, inertia estimation, event location and magnitude estimation, data authentication, control, and stability assessment. These applications are operating on a real-world system called FNET-GridEye, which is a wide-area measurement network and arguably the world-largest cyber-physical system that collects power grid big data. These applications showed much better performance compared with conventional approaches and accomplished new tasks that are impossible to realized using conventional technologies. These encouraging results demonstrate that combining power grid big data and artificial intelligence can uncover and capture the non-linear correlation between power grid data and its stabilities indices and will potentially enable many advanced applications that can significantly improve power grid resilience.

eess.SY↗

arXiv2021

Cell phenotypic transition proceeds through concerted reorganization of gene regulatory network

Phenotype transition takes place in many biological processes such as differentiation, and understanding how a cell reprograms its global gene expression profile is a problem of rate theories. A cell phenotype transition accompanies with switching of expression rates of clusters of genes, analogous to domain flipping in an Ising system. Here through analyzing single cell RNA sequencing data in the framework of transition path theory, we set to study how such a genome-wide expression program switching proceeds in three different cell transition processes. For each process after reconstructing a Markov transition model in the cell state space, we formed an ensemble of shortest paths connecting the initial and final cell states, reconstructed a reaction coordinate describing the transition progression, and inferred the gene regulation network (GRN) along the reaction coordinate. In all three processes we observed common pattern that the frustration of gene regulatory network (GRN), defined as overall confliction between the regulation received by genes and their expression states, first increases then decreases when approaching a new phenotype. The results support a mechanism of concerted silencing of genes that are active in the initial phenotype and activation of genes that are active in the final phenotype.

q-bio.MN↗

arXiv2021

Calculate Center-of-Inertia Frequency and System RoCoF Using PMU Data

The power system frequency is important for the system overall stability. However, there does not exist a single measurement point of the system frequency due to the distributed nature of the system inertia and the small inconsistency of different generator rotor electrical speeds in one synchronized system. This paper proposed a new approach to calculate the system center-of-inertia (COI) frequency and the rate-of-change-of-frequency (RoCoF) more accurately using PMU data at multiple locations. The COI frequency and the RoCoF value were further used to assist fast estimation of the imbalance MW amount of a frequency event. Test results using actual measurements in the U.S. Eastern Interconnection system validated the effectiveness of the proposed method.

eess.SY↗

arXiv2019

Are You for Real? Detecting Identity Fraud via Dialogue Interactions

Identity fraud detection is of great importance in many real-world scenarios such as the financial industry. However, few studies addressed this problem before. In this paper, we focus on identity fraud detection in loan applications and propose to solve this problem with a novel interactive dialogue system which consists of two modules. One is the knowledge graph (KG) constructor organizing the personal information for each loan applicant. The other is structured dialogue management that can dynamically generate a series of questions based on the personal KG to ask the applicants and determine their identity states. We also present a heuristic user simulator based on problem analysis to evaluate our method. Experiments have shown that the trainable dialogue system can effectively detect fraudsters, and achieve higher recognition accuracy compared with rule-based systems. Furthermore, our learned dialogue strategies are interpretable and flexible, which can help promote real-world applications.

cs.CL↗