arXiv subjects

Yun Hao

Publications and source records attributed to Yun Hao.

15 recordsLinked to original sources

arXiv2026

StyleForge: Indoor Furniture Styling by Counterfactual Reasoning in a Hypergraph Field

Fixed-layout indoor furniture styling requires selecting assets that form a coherent room without changing the prescribed furniture categories, positions, orientations, or scales. Existing approaches typically retrieve each asset independently or rely on static local relations, making them prone to shape, material, and color conflicts after scene composition. We introduce StyleForge, a scene-level structured selection framework built on a dynamic hypergraph style field. A frozen multimodal large language model extracts structured style priors from an open-ended style request and the fixed layout, while StyleForge maintains a learnable candidate distribution for each furniture slot. Conditioned on the target style, the dynamic hypergraph style field adaptively activates and weights layout-induced hyperedges to capture higher-order dependencies among furniture. Counterfactual style preference learning then treats each candidate as a local substitution in the current style field and evaluates its contextual compatibility using Mahalanobis energies. Training alternates between optimizing the style field and the candidate logits. At inference, the model remains frozen and test-time training updates only room-specific candidate logits, progressively correcting cross-slot style conflicts as the global scene context evolves. Experiments on 3D-FRONT demonstrate state-of-the-art furniture retrieval and scene-level style coherence, producing more coherent fixed-layout furniture arrangements than object- and scene-level retrieval baselines.

cs.CV↗

arXiv2026

Roomer: Reflective Object-Grounded Model Editing and Repair for 3D Indoor Layout Synthesis

Existing indoor layout generators produce globally plausible layouts yet may retain local violations such as collisions, out-of-bounds placements, obstructed openings, and blocked circulation. Most prior work focuses on full-scene synthesis or scene-level optimization, with limited support for identifying responsible objects and locally repairing affected regions. We present Roomer, a reflective repair framework that casts these violations as sparse, object-grounded repair problems. Roomer encodes layouts as ``RoState'' and uses ``RoReview'' to bind measured violations to implicated objects. A geometry-conditioned vision-language model planner proposes a structured local edit, while a deterministic solver validates it and generates a finite set of candidate edits when needed. Each candidate is committed only if full-scene verification confirms that it resolves the target violation without new hard violations or broken protected constraints. We train the planner on Roomer-CC, a controlled-corruption dataset that pairs faulty layouts with object-grounded violation evidence and known-feasible inverse StatePatches. Since existing benchmarks rarely assess whether physically valid layouts are usable, we introduce Roomer-Eval to assess distributional quality, physical validity, and practical usability. Experiments show that Roomer repairs residual violations while preserving valid regions, improves physical validity and usability, and transfers across external generators.

cs.RO↗

arXiv2026

Improving low-resource ASR using bilingual fine-tuning with language identification: a cross-linguistic evaluation

This study explores how bilingual fine-tuning affects automatic speech recognition (ASR) in low-resource languages. We evaluate this method across nine linguistically and geographically diverse language pairs, covering a range of language families and writing systems. To distinguish the two languages, during training, we pre-pend each input text with a language identification token. At inference, the model jointly predicts both the language and transcription from the speech input alone. As texts for which the language is incorrectly determined show low ASR performance, we also conduct a follow-up experiment in which the language identification token is provided both during training and inference. Our results show that bilingual fine-tuning can be beneficial when language identification accuracy is high, and that in cases where language identification performance is low, including the language identification token at inference helps to improve ASR performance.

cs.CL↗

arXiv2026

Can Large Language Models Reliably Correct Errors in Low-Resource ASR? A Contamination-Aware Case Study on West Frisian

Automatic speech recognition (ASR) has improved substantially in recent years, yet performance remains limited for low-resource languages. Large language models (LLMs) have shown promise for improving ASR through generative error correction (GER), but their effectiveness in low-resource settings remains underexplored. In addition, it remains unclear to what extent data contamination influences the reported improvements in LLM-based GER. This study investigates LLM-based GER for low-resource Frisian. In addition to a public corpus, we construct and use a Frisian offline dataset with non-public texts for evaluation to control for potential data contamination. Results show that GER improves ASR performance in most settings, with the best GPT-5.1 results surpassing oracle WERs. Comparable gains on the offline dataset indicate that improvements reflect true correction ability. We further provide a detailed error analysis revealing model correction patterns.

cs.CL↗

arXiv2025

Video Anomaly Detection with Semantics-Aware Information Bottleneck

Semi-supervised video anomaly detection methods face two critical challenges: (1) Strong generalization blurs the boundary between normal and abnormal patterns. Although existing approaches attempt to alleviate this issue using memory modules, their rigid prototype-matching process limits adaptability to diverse scenarios; (2) Relying solely on low-level appearance and motion cues makes it difficult to perceive high-level semantic anomalies in complex scenes. To address these limitations, we propose SIB-VAD, a novel framework based on adaptive information bottleneck filtering and semantic-aware enhancement. We propose the Sparse Feature Filtering Module (SFFM) to replace traditional memory modules. It compresses normal features directly into a low-dimensional manifold based on the information bottleneck principle and uses an adaptive routing mechanism to dynamically select the most suitable normal bottleneck subspace. Trained only on normal data, SFFMs only learn normal low-dimensional manifolds, while abnormal features deviate and are effectively filtered. Unlike memory modules, SFFM directly removes abnormal information and adaptively handles scene variations. To improve semantic awareness, we further design a multimodal prediction framework that jointly models appearance, motion, and semantics. Through multimodal consistency constraints and joint error computation, it achieves more robust VAD performance. Experimental results validate the effectiveness of our feature filtering paradigm based on semantics-aware information bottleneck. Project page at https://qzfm.github.io/sib_vad_project_page/

cs.CV↗

arXiv2025

Inverse Materials Design by Large Language Model-Assisted Generative Framework

Deep generative models hold great promise for inverse materials design, yet their efficiency and accuracy remain constrained by data scarcity and model architecture. Here, we introduce AlloyGAN, a closed-loop framework that integrates Large Language Model (LLM)-assisted text mining with Conditional Generative Adversarial Networks (CGANs) to enhance data diversity and improve inverse design. Taking alloy discovery as a case study, AlloyGAN systematically refines material candidates through iterative screening and experimental validation. For metallic glasses, the framework predicts thermodynamic properties with discrepancies of less than 8% from experiments, demonstrating its robustness. By bridging generative AI with domain knowledge and validation workflows, AlloyGAN offers a scalable approach to accelerate the discovery of materials with tailored properties, paving the way for broader applications in materials science.

cond-mat.mtrl-sci↗

arXiv2025

Evaluating Standard and Dialectal Frisian ASR: Multilingual Fine-tuning and Language Identification for Improved Low-resource Performance

Automatic Speech Recognition (ASR) performance for low-resource languages is still far behind that of higher-resource languages such as English, due to a lack of sufficient labeled data. State-of-the-art methods deploy self-supervised transfer learning where a model pre-trained on large amounts of data is fine-tuned using little labeled data in a target low-resource language. In this paper, we present and examine a method for fine-tuning an SSL-based model in order to improve the performance for Frisian and its regional dialects (Clay Frisian, Wood Frisian, and South Frisian). We show that Frisian ASR performance can be improved by using multilingual (Frisian, Dutch, English and German) fine-tuning data and an auxiliary language identification task. In addition, our findings show that performance on dialectal speech suffers substantially, and, importantly, that this effect is moderated by the elicitation approach used to collect the dialectal data. Our findings also particularly suggest that relying solely on standard language data for ASR evaluation may underestimate real-world performance, particularly in languages with substantial dialectal variation.

cs.CL↗

arXiv2024

A Multi-agent Framework for Physical Laws Discovery

Discovering explicit physical laws has traditionally depended on human intuition and domain expertise. Recent advances in artificial intelligence, particularly large language models (LLMs), offer a new route to accelerate this process by automating key steps from hypothesis generation to interpretable model construction. Here we develop an LLM-based multi-agent framework for physical-law discovery that integrates literature-guided variable selection, hypothesis formulation, symbolic regression, formula derivation, and mechanistic explanation. We validate the framework on three representative materials problems: the glass-forming ability (GFA) of metallic glasses, the Vickers hardness of compounds, and the Young's modulus of multi-component alloys. Using physically and chemically meaningful descriptors as inputs, the discovered formulas achieve strong agreement with reference data, with correlation coefficients up to 0.94 (GFA), 0.86 (hardness), and 0.94 (Young's modulus), while remaining compact and interpretable. Beyond fitting, the Young's modulus formula generalizes to quaternary and quinary alloys, improving prediction accuracy by up to 78.8% relative to the classical rule of mixtures. By integrating cross-disciplinary knowledge, reflection mechanisms, and expert-like reasoning ability into symbolic regression, our AI-centric framework offers a robust and extensible platform for automated physical laws discovery, demonstrating that AI can increasingly serve as an essential role in modern scientific research by thinking and acting like field experts.

cond-mat.mtrl-sci↗

arXiv2024

An Overview of zbMATH Open Digital Library

Mathematical research thrives on the effective dissemination and discovery of knowledge. zbMATH Open has emerged as a pivotal platform in this landscape, offering a comprehensive repository of mathematical literature. Beyond indexing and abstracting, it serves as a unified quality-assured infrastructure for finding, evaluating, and connecting mathematical information that advances mathematical research as well as interdisciplinary exploration. zbMATH Open enables scientific quality control by post-publication reviews and promotes connections between researchers, institutions, and research outputs. This paper represents the functionalities of the most significant features of this open-access service, highlighting its role in shaping the future of mathematical information retrieval.

cs.DL↗

arXiv2024

Hybrid Classification-Regression Adaptive Loss for Dense Object Detection

For object detection detectors, enhancing model performance hinges on the ability to simultaneously consider inconsistencies across tasks and focus on difficult-to-train samples. Achieving this necessitates incorporating information from both the classification and regression tasks. However, prior work tends to either emphasize difficult-to-train samples within their respective tasks or simply compute classification scores with IoU, often leading to suboptimal model performance. In this paper, we propose a Hybrid Classification-Regression Adaptive Loss, termed as HCRAL. Specifically, we introduce the Residual of Classification and IoU (RCI) module for cross-task supervision, addressing task inconsistencies, and the Conditioning Factor (CF) to focus on difficult-to-train samples within each task. Furthermore, we introduce a new strategy named Expanded Adaptive Training Sample Selection (EATSS) to provide additional samples that exhibit classification and regression inconsistencies. To validate the effectiveness of the proposed method, we conduct extensive experiments on COCO test-dev. Experimental evaluations demonstrate the superiority of our approachs. Additionally, we designed experiments by separately combining the classification and regression loss with regular loss functions in popular one-stage models, demonstrating improved performance.

cs.CV↗

arXiv2019

A Simpson correspondence for abelian varieties in characteristic p > 0

Let X/k be an abelian variety over an algebraically closed field k of characteristic p > 0. In this paper, using the Azumaya property of the sheaf of crystalline differential operators and the Morita equivalence, we show that etale locally over the Hitchin base, the moduli stack of Higgs bundles on the Frobenius twist X' is equivalent to that of local systems on X. We follow the approach of [Gro16].

math.AG↗

arXiv2019

Violation of $U_1$ Gauge Symmetry by Yang-Mills Gravity and Deflection of Light Experiment

Based on the gauge symmetry framework, the $U_1$ symmetry of electrodynamics is violated in the presence of gravity with space-time translational gauge symmetry in inertial frames. For a light ray, an eikonal equation with effective metric tensors is derived in the geometric-optics limit. Under these conditions, the angle of the deflection of light by the sun is calculated to be $\d ϕ\approx 1.75''$ in inertial frames without requiring a gauge condition such as $\p_μA^μ=0$. In contrast, if the theory is $U_1$ gauge invariant, one can impose the gauge condition $\p_μA^μ=0$ in the derivation of the eikonal equation. In this case, one obtains a slightly different effective metric tensor and a different angle of deflection $\d ϕ\approx 1.52''$. However, because the precision of experiments in the last century using optical frequencies has been no better than (10 $-$ 20)\% due to large systematic errors, one cannot unambiguously rule out the result $\d ϕ\approx 1.52'$. It is hoped that the precision of these data can be improved in order to test Yang-Mills gravity.

physics.gen-ph↗

arXiv2017

Integrating Knowledge from Latent and Explicit Features for Triple Scoring - Team Radicchio's Triple Scorer at WSDM Cup 2017

The objective of the triple scoring task in WSDM Cup 2017 is to compute relevance scores for knowledge-base triples of type-like relations. For example, consider Julius Caesar who has had various professions, including Politician and Author. For two given triples (Julius Caesar, profession, Politician) and (Julius Caesar, profession, Author), the former triple is likely to have a higher relevance score (also called "triple score") because Julius Caesar was well-known as a politician and not as an author. Accurate prediction of such triple scores greatly benefits real-world applications, such as information retrieval or knowledge base query. In these scenarios, being able to rank all relations (Profession/Nationality) can help improve the user experience. We propose a triple scoring model which integrates knowledge from both latent features and explicit features via an ensemble approach. The latent features consist of representations for a person learned by using a word2vec model and representations for profession/nationality values extracted from a pre-trained GloVe embedding model. In addition, we extract explicit features for person entities from the Freebase knowledge base. Experimental results show that the proposed method performs competitively at WSDM Cup 2017, ranking at the third place with an accuracy of 79.72% for predicting within two places of the ground truth score.

cs.IR↗

arXiv2016

Urban sidewalks: visualization and routing for individuals with limited mobility

People with limited mobility in the U.S. (defined as having difficulty or inability to walk a quarter of a mile without help and without the use of special equipment) face a growing informational gap: while pedestrian routing algorithms are getting faster and more informative, planning a route with a wheeled device in urban centers is very difficult due to lack of integrated pertinent information regarding accessibility along the route. Moreover, reducing access to street-spaces translates to reduced access to other public information and services that are increasingly made available to the public along urban streets. To adequately plan a commute, a traveler with limited or wheeled mobility must know whether her path may be blocked by construction, whether the sidewalk would be too steep or rendered unusable due to poor conditions, whether the street can be crossed or a highway is blocking the way, or whether there is a sidewalk at all. These details populate different datasets in many modern municipalities, but they are not immediately available in a convenient, integrated format to be useful to people with limited mobility. Our project, AccessMap, in its first phase (v.1) overlayed the information that is most relevant to people with limited mobility on a map, enabling self-planning of routes. Here, we describe the next phase of the project: synthesizing commonly available open data (including streets, sidewalks, curb ramps, elevation data, and construction permit information) to generate a graph of paths to enable variable cost-function accessible routing.

cs.CY↗

arXiv2014

Millimeter Wave and Terahertz Spectra and Global Fit of Torsion-Rotation Transitions in the Ground, First and Second Excited Torsional States of 13CH3OH Methanol

Methanol is observed in a wide range of astrophysical sources throughout the universe, and comprehensive databases of the millimeter and THz spectra of CH3OH and its principal isotopologues represent important tools for the astronomical community. A previous combined analysis of microwave and millimeter wave spectra of 13CH3OH together with Fourier transform far-infrared spectra was limited to the first two torsional states, v_t = 0 and 1, for J values up to 20. The limits on frequency and quantum number coverage have recently been extended by new millimeter and THz measurements on several different spectrometers in the Cologne laboratory in the frequency windows 34-70 GHz, 75-120 GHz, 240-340 GHz, 360-450 GHz and 1.12-1.50 THz. With the new data, the global treatment has now been expanded to include the first three torsional states for J values up to 30. The current 13CH3OH data set contains about 2,300 microwave, millimeter-wave, sub-millimeter and THz lines and about 17,100 Fourier-transform far-infrared lines, representing the most recent available information in the quantum number ranges J </= 30, K </= 13 and v_t </= 2. The transitions have been successfully fitted to within the assigned measurement uncertainties of +/-50 kHz for most of the frequency-measured (i.e. MW, MMW, Sub-MMW, THz) lines and +/-6 MHz for the FIR lines. A convergent global fit was achieved using 103 adjustable parameters to reach an overall weighted standard deviation of 1.37. Our new C-13 methanol database is improved substantially compared to the existing one (Li-Hong Xu, F.J. Lovas, J. Phys. Chem. Ref. Data 26 (1997) 17-156), and will be available in the Cologne Database for Molecular Spectroscopy, CDMS (http://www.astro.unikoeln.de/cdms/). (Abbreviated)

astro-ph.IM↗