arXiv subjects

Yihao Wang

Publications and source records attributed to Yihao Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

PAWS: Perception of Articulation in the Wild at Scale from Egocentric Videos

Articulation perception aims to recover the motion and structure of articulated objects (e.g., drawers and cupboards), and is fundamental to 3D scene understanding in robotics, simulation, and animation. Existing learning-based methods rely heavily on supervised training with high-quality 3D data and manual annotations, limiting scalability and diversity. To address this limitation, we propose PAWS, a method that directly extracts object articulations from hand-object interactions in large-scale in-the-wild egocentric videos. We evaluate our method on the public data sets, including HD-EPIC and Arti4D data sets, achieving significant improvements over baselines. We further demonstrate that the extracted articulations benefit downstream tasks, including fine-tuning 3D articulation prediction models and enabling robot manipulation. See the project website at https://aaltoml.github.io/PAWS/.

cs.CV↗

arXiv2026

ResAgent: Entropy-based Prior Point Discovery and Visual Reasoning for Referring Expression Segmentation

Referring Expression Segmentation (RES) is a core vision-language segmentation task that enables pixel-level understanding of targets via free-form linguistic expressions, supporting critical applications such as human-robot interaction and augmented reality. Despite the progress of Multimodal Large Language Model (MLLM)-based approaches, existing RES methods still suffer from two key limitations: first, the coarse bounding boxes from MLLMs lead to redundant or non-discriminative point prompts; second, the prevalent reliance on textual coordinate reasoning is unreliable, as it fails to distinguish targets from visually similar distractors. To address these issues, we propose \textbf{\model}, a novel RES framework integrating \textbf{E}ntropy-\textbf{B}ased Point \textbf{D}iscovery (\textbf{EBD}) and \textbf{V}ision-\textbf{B}ased \textbf{R}easoning (\textbf{VBR}). Specifically, EBD identifies high-information candidate points by modeling spatial uncertainty within coarse bounding boxes, treating point selection as an information maximization process. VBR verifies point correctness through joint visual-semantic alignment, abandoning text-only coordinate inference for more robust validation. Built on these components, \model implements a coarse-to-fine workflow: bounding box initialization, entropy-guided point discovery, vision-based validation, and mask decoding. Extensive evaluations on four benchmark datasets (RefCOCO, RefCOCO+, RefCOCOg, and ReasonSeg) demonstrate that \model achieves new state-of-the-art performance across all four benchmarks, highlighting its effectiveness in generating accurate and semantically grounded segmentation masks with minimal prompts.

cs.CV↗

arXiv2026

Subsystem fidelity in two-dimensional conformal field theories

We investigate the short-interval expansion of the subsystem fidelity in two-dimensional conformal field theories (2D CFTs) using the operator product expansion (OPE) of twist operators. We obtain universal contributions from general quasiprimary operators valid for arbitrary 2D CFTs, along with specific results in free massless boson and fermion theories. The analytical predictions demonstrate excellent agreement with established analytical results in field theories and numerical calculations in integrable models. Furthermore, we extend the method to holographic CFTs, where subsystem fidelity serves to analyze the distinguishability of black hole microstates through the AdS/CFT correspondence. This work establishes a unified framework for quantifying quantum state distinguishability across various 2D CFTs, bridging quantum information techniques with applications in quantum gravity.

hep-th↗

arXiv2025

Deterministic Electrical Control of Single Magnetic Bubbles in Nanostructured Cells

Localized particle-like spin textures have been found to exhibit emergent electromagnetic properties, which hold promise for the development of intriguing spintronic devices. Among these textures, magnetic bubbles represent localized spin configurations that could serve as data bits. However, the precise methods for their electrical manipulation remain uncertain. Here, we demonstrate the deterministic electrical manipulations and detections of single magnetic bubbles in kagome-latticed Fe3Sn2 magnetic nanostructured cells. The current-induced dynamics of magnetic bubbles were explored using nanosecond pulsed currents. We show single pulsed currents with low and high densities can be applied for the creation and deletion of a single bubble, respectively. The mutual writing-deleting operations on single bubbles are attributed to the thermal heating and non-thermal spin-transfer torque effects in combination with micromagnetic simulations. We also realized the in-situ detection of a single bubble using the anisotropic magnetoresistance effect through a standard four-probe method. Our results could propel the development of bubble-based spintronic devices.

cond-mat.mes-hall↗

arXiv2025

Creating and Deleting a Single Dipolar Skyrmion by Surface Spin Twists

We report deterministic operations on single dipolar skyrmions confined in nanostructured cuboids using in-plane currents. We achieve highly reversible writing and deleting of skyrmions in the simple cuboid without any artificial defects or pinning sites. The current-induced creation of skyrmions is well-understood through the spin-transfer torque acting on surface spin twists of the spontaneous 3D ferromagnetic state, caused by the magnetic dipole-dipole interaction of the uniaxial Fe3Sn2 magnet with a low-quality factor. Current-induced deletions of skyrmions result from the combined effects of magnetic hysteresis and Joule thermal heating. Our results are replicated consistently through 3D micromagnetic simulations. Our approach offers a viable method for achieving reliable single-bit operations in skyrmionic devices for applications such as random-access memories.

cond-mat.mes-hall↗

arXiv2025

On the Potential of Digital Twins for Distribution System State Estimation with Randomly Missing Data in Heterogeneous Measurements

Traditional statistical optimization-based state estimation (DSSE) algorithms rely on detailed grid parameters and mathematical assumptions of all possible uncertainties. Furthermore, random data missing due to communication failures, congestion, and cyberattacks, makes these methods easily infeasible. Inspired by recent advances in digital twins (DTs), this paper proposes an interactive attention-based DSSE model for robust grid monitoring by integrating three core components: physical entities, virtual modeling, and data fusion. To enable robustness against various data missing in heterogeneous measurements, we first propose physics-informed data augmentation and transfer. Moreover, a state-of-the-art attention-based spatiotemporal feature learning is proposed, followed by a novel cross-interaction feature fusion for robust voltage estimation. A case study in a real-world unbalanced 84-bus distribution system with raw data validates the accuracy and robustness of the proposed DT model in estimating voltage states, with random locational, arbitrary ratios (up to 40% of total measurements) of data missing.

eess.SY↗

arXiv2025

LIAM: Multimodal Transformer for Language Instructions, Images, Actions and Semantic Maps

The availability of large language models and open-vocabulary object perception methods enables more flexibility for domestic service robots. The large variability of domestic tasks can be addressed without implementing each task individually by providing the robot with a task description along with appropriate environment information. In this work, we propose LIAM - an end-to-end model that predicts action transcripts based on language, image, action, and map inputs. Language and image inputs are encoded with a CLIP backbone, for which we designed two pre-training tasks to fine-tune its weights and pre-align the latent spaces. We evaluate our method on the ALFRED dataset, a simulator-generated benchmark for domestic tasks. Our results demonstrate the importance of pre-aligning embedding spaces from different modalities and the efficacy of incorporating semantic maps.

cs.CV↗

arXiv2025

VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulators

Vision-Language-Action (VLA) models enable embodied decision-making but rely heavily on imitation learning, leading to compounding errors and poor robustness under distribution shift. Reinforcement learning (RL) can mitigate these issues yet typically demands costly real-world interactions or suffers from sim-to-real gaps. We introduce VLA-RFT, a reinforcement fine-tuning framework that leverages a data-driven world model as a controllable simulator. Trained from real interaction data, the simulator predicts future visual observations conditioned on actions, allowing policy rollouts with dense, trajectory-level rewards derived from goal-achieving references. This design delivers an efficient and action-aligned learning signal, drastically lowering sample requirements. With fewer than 400 fine-tuning steps, VLA-RFT surpasses strong supervised baselines and achieves greater efficiency than simulator-based RL. Moreover, it exhibits strong robustness under perturbed conditions, sustaining stable task execution. Our results establish world-model-based RFT as a practical post-training paradigm to enhance the generalization and robustness of VLA models. For more details, please refer to https://vla-rft.github.io/.

cs.RO↗

arXiv2025

VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model

Vision-Language-Action (VLA) models typically bridge the gap between perceptual and action spaces by pre-training a large-scale Vision-Language Model (VLM) on robotic data. While this approach greatly enhances performance, it also incurs significant training costs. In this paper, we investigate how to effectively bridge vision-language (VL) representations to action (A). We introduce VLA-Adapter, a novel paradigm designed to reduce the reliance of VLA models on large-scale VLMs and extensive pre-training. To this end, we first systematically analyze the effectiveness of various VL conditions and present key findings on which conditions are essential for bridging perception and action spaces. Based on these insights, we propose a lightweight Policy module with Bridge Attention, which autonomously injects the optimal condition into the action space. In this way, our method achieves high performance using only a 0.5B-parameter backbone, without any robotic data pre-training. Extensive experiments on both simulated and real-world robotic benchmarks demonstrate that VLA-Adapter not only achieves state-of-the-art level performance, but also offers the fast inference speed reported to date. Furthermore, thanks to the proposed advanced bridging paradigm, VLA-Adapter enables the training of a powerful VLA model in just 8 hours on a single consumer-grade GPU, greatly lowering the barrier to deploying the VLA model. Project page: https://vla-adapter.github.io/.

cs.RO↗

arXiv2025

Pairwise correlations of global times in one-dimensional Brownian motion under stochastic resetting

Brownian motion with stochastic resetting-a process combining standard diffusion with random returns to a fixed position-has emerged as a powerful framework with applications spanning statistical physics, chemical kinetics, biology, and finance. In this study, we investigate the mutual correlations among three global characteristic times for one-dimensional resetting Brownian motion $x(τ)$ over the interval $τ\in \left[ 0, t\right] $: the occupation time $t_o$ spent on the positive semi-axis, the time $t_m$ at which $x(τ)$ attains its global maximum, and the last-passage time $t_{\ell}$ when the process crosses the origin. For the process starting from the origin and undergoing Poissonian resetting back to the origin, we analytically compute the pairwise joint distributions of these three times (in the Laplace domain) and derive their pairwise correlation coefficients. Our results reveal that these global times display rich correlations, with a non-trivial dependence on the resetting rate $r$. Specifically, we find that (i) While $t_{o}$ and $t_{\ell}^m$ are uncorrelated for any positive integer $m$, $t_{o}^2$ and $t_{\ell}^m$ display anti-correlation; (ii) A positive correlation exists between $t_{o}$ and $t_{m}$, which decays toward zero following a logarithmically corrected power-law way with an exponent of $-2$ as $r \to \infty$; (iii) The correlation between $t_{m}$ and $t_{\ell}$ shifts from positive to negative as $r$ increases. All analytical predictions are validated by extensive numerical simulations.

cond-mat.stat-mech↗

arXiv2025

Uncalibrated Cosmic Standards as a Robust Test on Late-Time Cosmological Models

We present an assumption-minimized framework for testing late-time cosmological models using Uncalibrated Cosmic Standards (UCS), including standard rulers and standard candles, without relying on absolute calibrations. The method exploits a tight, model-insensitive correlation between the sound horizons at recombination and the drag epoch. By avoiding dependence on pre-recombination physics and the amplitude of the Cosmic Microwave Background (CMB) power spectra, the UCS framework reduces potential early-Universe biases while retaining much of the constraining power of full analyses. Applying UCS to the recent dynamical dark energy (DE) study that reported deviations from $Λ$CDM, we find the constraints shift systematically toward the $Λ$CDM case. If this shift is physical, it may result from the omission of some pre-recombination physical processes that influence the scale dependence of the CMB spectra. We also observe a mild tension between uncalibrated standard rulers and candles, which can be largely mitigated by introducing a redshift-dependent magnitude bias in the supernova (SNe Ia) data. Our results highlight the importance of isolating post-recombination observables for testing late-time models in the era of precision cosmology, positioning UCS analysis as a robust framework for upcoming galaxy surveys.

astro-ph.CO↗

arXiv2025

Quantum Knowledge Distillation for Large Language Models

As foundational tools in natural language processing, Large Language Models (LLMs) have immense parameter scales, which makes deployment and inference increasingly prohibitive, especially in resource-constrained devices. Therefore, knowledge distillation for LLMs, i.e., compressing the LLM to a smaller model, is meaningful. With strong parameter representation capacity, quantum computing is regarded as a promising solution. Here, we propose a Quantum knowledge Distillation model for LLMs (QD-LLM) that leverages variational quantum circuits to learn from LLMs. In classical simulation, QD-LLM outperforms several mainstream distillation methods on multiple text classification tasks in terms of both accuracy and efficiency using only 11 qubits. The results reveal an interesting phenomenon that the simulation of quantum student models may be regarded as a new class of quantum-inspired classical algorithms. Remarkably, we deploy the obtained circuits on the Baihua superconducting quantum processor via the Quafu platform to assess practical feasibility. The model maintains stable inference performance despite hardware constraints such as decoherence and finite sampling. In summary, QD-LLM marks a foundational step in connecting quantum computing with LLMs, demonstrating the feasibility of quantum-native approaches that aim to compress and deploy models of increasingly larger scales. The code of this article has been open-sourced at https://github.com/Lilingxiao-bupt/QD-LLM.

quant-ph↗

arXiv2025

FMNV: A Dataset of Media-Published News Videos for Fake News Detection

News media, particularly video-based platforms, have become deeply embed-ded in daily life, concurrently amplifying the risks of misinformation dissem-ination. Consequently, multimodal fake news detection has garnered signifi-cant research attention. However, existing datasets predominantly comprise user-generated videos characterized by crude editing and limited public en-gagement, whereas professionally crafted fake news videos disseminated by media outlets-often politically or virally motivated-pose substantially greater societal harm. To address this gap, we construct FMNV, a novel da-taset exclusively composed of news videos published by media organizations. Through empirical analysis of existing datasets and our curated collection, we categorize fake news videos into four distinct types. Building upon this taxonomy, we employ Large Language Models (LLMs) to automatically generate deceptive content by manipulating authentic media-published news videos. Furthermore, we propose FMNVD, a baseline model featuring a dual-stream architecture that integrates spatio-temporal motion features from a 3D ResNeXt-101 backbone and static visual semantics from CLIP. The two streams are fused via an attention-based mechanism, while co-attention modules refine the visual, textual, and audio features for effective multi-modal aggregation. Comparative experiments demonstrate both the generali-zation capability of FMNV across multiple baselines and the superior detec-tion efficacy of FMNVD. This work establishes critical benchmarks for de-tecting high-impact fake news in media ecosystems while advancing meth-odologies for cross-modal inconsistency analysis. Our dataset is available in https://github.com/DennisIW/FMNV.

cs.CV↗

arXiv2025

Dynamically Allocated Interval-Based Generative Linguistic Steganography with Roulette Wheel

Existing linguistic steganography schemes often overlook the conditional probability (CP) of tokens in the candidate pool, allocating the one coding to all tokens, which results in identical selection likelihoods. This approach leads to the selection of low-CP tokens, degrading the quality of stegos and making them more detectable. This paper proposes a scheme based on the interval allocated, called DAIRstega. DAIRstega first uses a portion of the read secret to build the roulette area. Then, this scheme uses the idea of the roulette wheel and takes the CPs of tokens as the main basis for allocating the roulette area (i.e., the interval length). Thus, tokens with larger CPs are allocated more area. The secret will have an increased likelihood of selecting a token with a higher CP. During allocation, we designed some allocation functions and three constraints to optimize the process. Additionally, DAIRstega supports prompt-based controllable generation of stegos. Rich experiments show that the proposed embedding way and DAIRstega perform better than the existing ways and baselines, which shows strong perceptual, statistical, and semantic concealment, as well as anti-steganalysis ability. It can also generate high-quality longer stegos, addressing the deficiencies in this task. DAIRstega is confirmed to have potential as a secure watermarking, offering insights for its development.

cs.CL↗

arXiv2025

DeSplat: Decomposed Gaussian Splatting for Distractor-Free Rendering

Gaussian splatting enables fast novel view synthesis in static 3D environments. However, reconstructing real-world environments remains challenging as distractors or occluders break the multi-view consistency assumption required for accurate 3D reconstruction. Most existing methods rely on external semantic information from pre-trained models, introducing additional computational overhead as pre-processing steps or during optimization. In this work, we propose a novel method, DeSplat, that directly separates distractors and static scene elements purely based on volume rendering of Gaussian primitives. We initialize Gaussians within each camera view for reconstructing the view-specific distractors to separately model the static 3D scene and distractors in the alpha compositing stages. DeSplat yields an explicit scene separation of static elements and distractors, achieving comparable results to prior distractor-free approaches without sacrificing rendering speed. We demonstrate DeSplat's effectiveness on three benchmark data sets for distractor-free novel view synthesis. See the project website at https://aaltoml.github.io/desplat/.

cs.CV↗

arXiv2025

A Stronger Mixture of Low-Rank Experts for Fine-Tuning Foundation Models

In order to streamline the fine-tuning of foundation models, Low-Rank Adapters (LoRAs) have been substantially adopted across various fields, including instruction tuning and domain adaptation. The underlying concept of LoRA involves decomposing a full-rank matrix into the product of two lower-rank matrices, which reduces storage consumption and accelerates the training process. Furthermore, to address the limited expressive capacity of LoRA, the Mixture-of-Expert (MoE) has been introduced for incorporating multiple LoRA adapters. The integration of LoRA experts leads to a visible improvement across several downstream scenes. However, the mixture of LoRAs (MoE-LoRA) still exhibits its low robustness during tuning and inferring. Inspired by the Riemannian Preconditioners which train LoRA as a sub-space projector, we propose a new training strategy for MoE-LoRA, to stabilize and boost its feature learning procedure by multi-space projections. Examinations on SGD and AdamW optimizers demonstrate the effectiveness of our methodology. Source code is available at https://github.com/THUDM/MoELoRA_Riemannian.

cs.LG↗

arXiv2025

Toward Copyright Integrity and Verifiability via Multi-Bit Watermarking for Intelligent Transportation Systems

Intelligent transportation systems (ITS) use advanced technologies such as artificial intelligence to significantly improve traffic flow management efficiency, and promote the intelligent development of the transportation industry. However, if the data in ITS is attacked, such as tampering or forgery, it will endanger public safety and cause social losses. Therefore, this paper proposes a watermarking that can verify the integrity of copyright in response to the needs of ITS, termed ITSmark. ITSmark focuses on functions such as extracting watermarks, verifying permission, and tracing tampered locations. The scheme uses the copyright information to build the multi-bit space and divides this space into multiple segments. These segments will be assigned to tokens. Thus, the next token is determined by its segment which contains the copyright. In this way, the obtained data contains the custom watermark. To ensure the authorization, key parameters are encrypted during copyright embedding to obtain cipher data. Only by possessing the correct cipher data and private key, can the user entirely extract the watermark. Experiments show that ITSmark surpasses baseline performances in data quality, extraction accuracy, and unforgeability. It also shows unique capabilities of permission verification and tampered location tracing, which ensures the security of extraction and the reliability of copyright verification. Furthermore, ITSmark can also customize the watermark embedding position and proportion according to user needs, making embedding more flexible.

cs.CR↗

arXiv2025

U-GIFT: Uncertainty-Guided Firewall for Toxic Speech in Few-Shot Scenario

With the widespread use of social media, user-generated content has surged on online platforms. When such content includes hateful, abusive, offensive, or cyberbullying behavior, it is classified as toxic speech, posing a significant threat to the online ecosystem's integrity and safety. While manual content moderation is still prevalent, the overwhelming volume of content and the psychological strain on human moderators underscore the need for automated toxic speech detection. Previously proposed detection methods often rely on large annotated datasets; however, acquiring such datasets is both costly and challenging in practice. To address this issue, we propose an uncertainty-guided firewall for toxic speech in few-shot scenarios, U-GIFT, that utilizes self-training to enhance detection performance even when labeled data is limited. Specifically, U-GIFT combines active learning with Bayesian Neural Networks (BNNs) to automatically identify high-quality samples from unlabeled data, prioritizing the selection of pseudo-labels with higher confidence for training based on uncertainty estimates derived from model predictions. Extensive experiments demonstrate that U-GIFT significantly outperforms competitive baselines in few-shot detection scenarios. In the 5-shot setting, it achieves a 14.92\% performance improvement over the basic model. Importantly, U-GIFT is user-friendly and adaptable to various pre-trained language models (PLMs). It also exhibits robust performance in scenarios with sample imbalance and cross-domain settings, while showcasing strong generalization across various language applications. We believe that U-GIFT provides an efficient solution for few-shot toxic speech detection, offering substantial support for automated content moderation in cyberspace, thereby acting as a firewall to promote advancements in cybersecurity.

cs.SD↗