arXiv subjects

Tao Zhang

Publications and source records attributed to Tao Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2025

CoSineVerifier: Tool-Augmented Answer Verification for Computation-Oriented Scientific Questions

Answer verification methods are widely employed in language model training pipelines spanning data curation, evaluation, and reinforcement learning with verifiable rewards (RLVR). While prior work focus on developing unified verifiers applicable across multiple reasoning scenarios, significant challenges remain in computation-oriented scientific domains, such as algebraic equivalence checking and physical constant substitution. In this paper, we introduce \model, a tool-augmented verifier that leverages external executors to perform precise computations and symbolic simplifications. \model enables robust verification that goes beyond simple semantic matching. We propose a novel two-stage pipeline, which begin with cold-start fine-tuning and followed by multi-turn reinforcement learning with tool integration. Extensive experiments conducted on STEM subjects, general QA, and long-form reasoning tasks demonstrates strong generalization of \model. The results shows that the \model achieves state-of-the-art performance on VerifyBench-Hard and SCI-Bench. And we also employ our \model in RLVR as a reward model, the results show that it consistently outperforms both rubric-based and model-based verifiers on AIME'24 and AIME'25, demonstrating strong potential to enhance reasoning capabilities of LLM. Our model is released at \hyperlink{https://huggingface.co/Nanbeige/CoSineVerifier-Tool-4B}{https://huggingface.co/Nanbeige/CoSineVerifier-Tool-4B}.

cs.LG↗

arXiv2025

Tomographic beta-gamma spectroscopy of nuclear beta decay

Nuclear $β$ decay, a sensitive probe of nuclear structure and weak interactions, has become a precision test bed for physics beyond the Standard Model, driven by recent advances in spectrometric techniques. Here we introduce tomographic $β$-$γ$ spectroscopy (TBGS) of nuclear $β$ decay, a method that detects the energies of $β$, $γ$, and internal conversion electrons while simultaneously reconstructing the energy deposition vertices. Using the PandaX-4T detector operated as a TBGS, we obtain a precise and unbiased decay scheme of $^{214}$Pb, a key background isotope in searches for dark matter and Majorana neutrinos. For the first time, transitions of $^{214}$Pb to both the ground and excited states of $^{214}$Bi are measured concurrently, revealing discrepancies in branching ratios of up to 4.7$σ$ relative to previous evaluations. Combined with state-of-the-art theoretical spectral-shape calculations, these results establish a new benchmark for background modelling in rare-event searches and highlight the potential of TBGS as a versatile tool for fundamental physics and nuclear applications.

nucl-ex↗

arXiv2025

DenseScan: Advancing 3D Scene Understanding with 2D Dense Annotation

3D understanding is a key capability for real-world AI assistance. High-quality data plays an important role in driving the development of the 3D understanding community. Current 3D scene understanding datasets often provide geometric and instance-level information, yet they lack the rich semantic annotations necessary for nuanced visual-language tasks.In this work, we introduce DenseScan, a novel dataset with detailed multi-level descriptions generated by an automated pipeline leveraging multi-view 2D images and multimodal large language models (MLLMs). Our approach enables dense captioning of scene elements, ensuring comprehensive object-level descriptions that capture context-sensitive details. Furthermore, we extend these annotations through scenario-based question generation, producing high-level queries that integrate object properties, spatial relationships, and scene context. By coupling geometric detail with semantic richness, DenseScan broadens the range of downstream tasks, from detailed visual-language navigation to interactive question answering. Experimental results demonstrate that our method significantly enhances object-level understanding and question-answering performance in 3D environments compared to traditional annotation pipelines. We release both the annotated dataset and our annotation pipeline to facilitate future research and applications in robotics, augmented reality, and beyond. Through DenseScan, we aim to catalyze new avenues in 3D scene understanding, allowing researchers and practitioners to tackle the complexities of real-world environments with richer, more contextually aware annotations.

cs.CV↗

arXiv2025

Non-abelian extensions and automorphisms of post-Lie algebras

In this paper, we introduce the concepts of crossed modules of post-Lie algebras and cat$^1$-post-Lie algebras. It is proved that these two concepts are equivalent to each other. Secondly, we construct a non-abelian cohomology for post-Lie algebras to classify their non-abelian extensions. At last, we investigate the inducibility problem of a pair of automorphisms for post-Lie algebras and construct a Wells exact sequence to solve it.

math.RA↗

arXiv2025

Wavefront-Constrained Passive Obscured Object Detection

Accurately localizing and segmenting obscured objects from faint light patterns beyond the field of view is highly challenging due to multiple scattering and medium-induced perturbations. Most existing methods, based on real-valued modeling or local convolutional operations, are inadequate for capturing the underlying physics of coherent light propagation. Moreover, under low signal-to-noise conditions, these methods often converge to non-physical solutions, severely compromising the stability and reliability of the observation. To address these challenges, we propose a novel physics-driven Wavefront Propagating Compensation Network (WavePCNet) to simulate wavefront propagation and enhance the perception of obscured objects. This WavePCNet integrates the Tri-Phase Wavefront Complex-Propagation Reprojection (TriWCP) to incorporate complex amplitude transfer operators to precisely constrain coherent propagation behavior, along with a momentum memory mechanism to effectively suppress the accumulation of perturbations. Additionally, a High-frequency Cross-layer Compensation Enhancement is introduced to construct frequency-selective pathways with multi-scale receptive fields and dynamically model structural consistency across layers, further boosting the model's robustness and interpretability under complex environmental conditions. Extensive experiments conducted on four physically collected datasets demonstrate that WavePCNet consistently outperforms state-of-the-art methods across both accuracy and robustness.

cs.CV↗

arXiv2025

The First Scientific Flight and Observations of the 50-mm Balloon-Borne White-Light Coronagraph

A 50-mm balloon-borne white-light coronagraph (BBWLC) to observe whitelight solar corona over the altitude range from 1.08 to 1.50 solar radii has recently been indigenously developed by Yunnan Observatories in collaboration with Shangdong University (in Weihai) and Changchun Institute of Optics, Fine Mechanics and Physics, which will significantly improve the ability of China to detect and measure inner corona. On 2022 October 4, its first scientific flight took place at the Dachaidan area in Qinghai province of China. We describe briefly the BBWLC mission including its optical design, mechanical structure, pointing system, the first flight and results associated with the data processing approach. Preliminary analysis of the data shows that BBWLC imaged the Kcorona with three streamer structures on the west limb of the Sun. To further confirm the coronal signals obtained by BBWLC, comparisonswere made with observations of the Kcoronagraph of the High Altitude Observatory and the Atmospheric ImagingAssembly on board the Solar Dynamics Observatory. We conclude that BBWLC eventually observed the white-light corona in its first scientific flight.

astro-ph.SR↗

arXiv2025

The singular anisotropic Adams' type inequality in $\mathbb{R}^n$

In this paper, using anisotropic rearrangement techniques, we first establish the best constants for the singular anisotropic Adams' type inequality with exact growth in $\mathbb{R}^n$. Furthermore, by the same trick, we also prove the singular anisotropic Adams' type inequality on bounded domain $Ω\subset \mathbb{R}^n$.

math.AP↗

arXiv2025

Head Stabilization for Wheeled Bipedal Robots via Force-Estimation-Based Admittance Control

Wheeled bipedal robots are emerging as flexible platforms for field exploration. However, head instability induced by uneven terrain can degrade the accuracy of onboard sensors or damage fragile payloads. Existing research primarily focuses on stabilizing the mobile platform but overlooks active stabilization of the head in the world frame, resulting in vertical oscillations that undermine overall stability. To address this challenge, we developed a model-based ground force estimation method for our 6-degree-of-freedom wheeled bipedal robot. Leveraging these force estimates, we implemented an admittance control algorithm to enhance terrain adaptability. Simulation experiments validated the real-time performance of the force estimator and the robot's robustness when traversing uneven terrain.

cs.RO↗

arXiv2025

Real Noise Decoupling for Hyperspectral Image Denoising

Hyperspectral image (HSI) denoising is a crucial step in enhancing the quality of HSIs. Noise modeling methods can fit noise distributions to generate synthetic HSIs to train denoising networks. However, the noise in captured HSIs is usually complex and difficult to model accurately, which significantly limits the effectiveness of these approaches. In this paper, we propose a multi-stage noise-decoupling framework that decomposes complex noise into explicitly modeled and implicitly modeled components. This decoupling reduces the complexity of noise and enhances the learnability of HSI denoising methods when applied to real paired data. Specifically, for explicitly modeled noise, we utilize an existing noise model to generate paired data for pre-training a denoising network, equipping it with prior knowledge to handle the explicitly modeled noise effectively. For implicitly modeled noise, we introduce a high-frequency wavelet guided network. Leveraging the prior knowledge from the pre-trained module, this network adaptively extracts high-frequency features to target and remove the implicitly modeled noise from real paired HSIs. Furthermore, to effectively eliminate all noise components and mitigate error accumulation across stages, a multi-stage learning strategy, comprising separate pre-training and joint fine-tuning, is employed to optimize the entire framework. Extensive experiments on public and our captured datasets demonstrate that our proposed framework outperforms state-of-the-art methods, effectively handling complex real-world noise and significantly enhancing HSI quality.

cs.CV↗

arXiv2025

From Narrow Unlearning to Emergent Misalignment: Causes, Consequences, and Containment in LLMs

Recent work has shown that fine-tuning on insecure code data can trigger an emergent misalignment (EMA) phenomenon, where models generate malicious responses even to prompts unrelated to the original insecure code-writing task. Such cross-domain generalization of harmful behavior underscores the need for a deeper understanding of the algorithms, tasks, and datasets that induce emergent misalignment. In this work, we extend this study by demonstrating that emergent misalignment can also arise from narrow refusal unlearning in specific domains. We perform refusal unlearning on Cybersecurity and Safety concept, and evaluate EMA by monitoring refusal scores across seven responsible AI (RAI) domains, Cybersecurity, Safety, Toxicity, Bias, Sensitive Content, Medical/Legal, and Privacy. Our work shows that narrow domain unlearning can yield compliance responses for the targeted concept, however, it may also propagate EMA to unrelated domains. Among the two intervened concepts, Cybersecurity and Safety, we find that the safety concept can have larger EMA impact, i.e, causing lower refusal scores, across other unrelated domains such as bias. We observe this effect consistently across two model families, Mistral-7b-0.3v, and Qwen-7b-2.5. Further, we show that refusal unlearning augmented with cross-entropy loss function on a small set of retain data from the affected domains can largely, if not fully, restore alignment across the impacted domains while having lower refusal rate on the concept we perform unlearning on. To investigate the underlying causes of EMA, we analyze concept entanglements at the representation level via concept vectors. Our analysis reveals that concepts with higher representation similarity in earlier layers are more susceptible to EMA after intervention when the refusal stream is altered through targeted refusal unlearning.

cs.LG↗

arXiv2025

FengHuang: Next-Generation Memory Orchestration for AI Inferencing

This document presents a vision for a novel AI infrastructure design that has been initially validated through inference simulations on state-of-the-art large language models. Advancements in deep learning and specialized hardware have driven the rapid growth of large language models (LLMs) and generative AI systems. However, traditional GPU-centric architectures face scalability challenges for inference workloads due to limitations in memory capacity, bandwidth, and interconnect scaling. To address these issues, the FengHuang Platform, a disaggregated AI infrastructure platform, is proposed to overcome memory and communication scaling limits for AI inference. FengHuang features a multi-tier shared-memory architecture combining high-speed local memory with centralized disaggregated remote memory, enhanced by active tensor paging and near-memory compute for tensor operations. Simulations demonstrate that FengHuang achieves up to 93% local memory capacity reduction, 50% GPU compute savings, and 16x to 70x faster inter-GPU communication compared to conventional GPU scaling. Across workloads such as GPT-3, Grok-1, and QWEN3-235B, FengHuang enables up to 50% GPU reductions while maintaining end-user performance, offering a scalable, flexible, and cost-effective solution for AI inference infrastructure. FengHuang provides an optimal balance as a rack-level AI infrastructure scale-up solution. Its open, heterogeneous design eliminates vendor lock-in and enhances supply chain flexibility, enabling significant infrastructure and power cost reductions.

cs.DC↗

arXiv2025

Tunable Hybrid-Mode Coupler Enabling Strong Interactions between Transmons at Centimeter-Scale Distance

The transmon, a fabrication-friendly superconducting qubit, remains a leading candidate for scalable quantum computing. Recent advances in tunable couplers have accelerated progress toward high-performance quantum processors. However, extending coherent interactions beyond millimeter scales to enhance quantum connectivity presents a critical challenge. Here, we introduce a hybrid-mode coupler exploiting resonator-transmon hybridization to simultaneously engineer the two lowest-frequency mode, enabling high-contrast coupling between centimeter-scale transmons. For a 1-cm coupler, our framework predicts flux-tunable $XX$ and $ZZ$ coupling strengths reaching 23 MHz and 100 MHz, with modulation contrasts exceeding $10^2$ and $10^4$, respectively, demonstrating quantitative agreement with an effective two-channel model. This work provides an efficient pathway to mitigate the inherent connectivity constraints imposed by short-range interactions, enabling transmon-based architectures compatible with hardware-efficient quantum tasks.

quant-ph↗

arXiv2025

Whole-Body Control With Terrain Estimation of A 6-DoF Wheeled Bipedal Robot

Wheeled bipedal robots have garnered increasing attention in exploration and inspection. However, most research simplifies calculations by ignoring leg dynamics, thereby restricting the robot's full motion potential. Additionally, robots face challenges when traversing uneven terrain. To address the aforementioned issue, we develop a complete dynamics model and design a whole-body control framework with terrain estimation for a novel 6 degrees of freedom wheeled bipedal robot. This model incorporates the closed-loop dynamics of the robot and a ground contact model based on the estimated ground normal vector. We use a LiDAR inertial odometry framework and improved Principal Component Analysis for terrain estimation. Task controllers, including PD control law and LQR, are employed for pose control and centroidal dynamics-based balance control, respectively. Furthermore, a hierarchical optimization approach is used to solve the whole-body control problem. We validate the performance of the terrain estimation algorithm and demonstrate the algorithm's robustness and ability to traverse uneven terrain through both simulation and real-world experiments.

cs.RO↗

arXiv2025

M3PD Dataset: Dual-view Photoplethysmography (PPG) Using Front-and-rear Cameras of Smartphones in Lab and Clinical Settings

Portable physiological monitoring is essential for early detection and management of cardiovascular disease, but current methods often require specialized equipment that limits accessibility or impose impractical postures that patients cannot maintain. Video-based photoplethysmography on smartphones offers a convenient noninvasive alternative, yet it still faces reliability challenges caused by motion artifacts, lighting variations, and single-view constraints. Few studies have demonstrated reliable application to cardiovascular patients, and no widely used open datasets exist for cross-device accuracy. To address these limitations, we introduce the M3PD dataset, the first publicly available dual-view mobile photoplethysmography dataset, comprising synchronized facial and fingertip videos captured simultaneously via front and rear smartphone cameras from 60 participants (including 47 cardiovascular patients). Building on this dual-view setting, we further propose F3Mamba, which fuses the facial and fingertip views through Mamba-based temporal modeling. The model reduces heart-rate error by 21.9 to 30.2 percent over existing single-view baselines while improving robustness in challenging real-world scenarios. Data and code: https://github.com/Health-HCI-Group/F3Mamba.

cs.CV↗

arXiv2025

No-Human in the Loop: Agentic Evaluation at Scale for Recommendation

Evaluating large language models (LLMs) as judges is increasingly critical for building scalable and trustworthy evaluation pipelines. We present ScalingEval, a large-scale benchmarking study that systematically compares 36 LLMs, including GPT, Gemini, Claude, and Llama, across multiple product categories using a consensus-driven evaluation protocol. Our multi-agent framework aggregates pattern audits and issue codes into ground-truth labels via scalable majority voting, enabling reproducible comparison of LLM evaluators without human annotation. Applied to large-scale complementary-item recommendation, the benchmark reports four key findings: (i) Anthropic Claude 3.5 Sonnet achieves the highest decision confidence; (ii) Gemini 1.5 Pro offers the best overall performance across categories; (iii) GPT-4o provides the most favorable latency-accuracy-cost tradeoff; and (iv) GPT-OSS 20B leads among open-source models. Category-level analysis shows strong consensus in structured domains (Electronics, Sports) but persistent disagreement in lifestyle categories (Clothing, Food). These results establish ScalingEval as a reproducible benchmark and evaluation protocol for LLMs as judges, with actionable guidance on scaling, reliability, and model family tradeoffs.

cs.AI↗

arXiv2025

MapSAM2: Adapting SAM2 for Automatic Segmentation of Historical Map Images and Time Series

Historical maps are unique and valuable archives that document geographic features across different time periods. However, automated analysis of historical map images remains a significant challenge due to their wide stylistic variability and the scarcity of annotated training data. Constructing linked spatio-temporal datasets from historical map time series is even more time-consuming and labor-intensive, as it requires synthesizing information from multiple maps. Such datasets are essential for applications such as dating buildings, analyzing the development of road networks and settlements, studying environmental changes etc. We present MapSAM2, a unified framework for automatically segmenting both historical map images and time series. Built on a visual foundation model, MapSAM2 adapts to diverse segmentation tasks with few-shot fine-tuning. Our key innovation is to treat both historical map images and time series as videos. For images, we process a set of tiles as a video, enabling the memory attention mechanism to incorporate contextual cues from similar tiles, leading to improved geometric accuracy, particularly for areal features. For time series, we introduce the annotated Siegfried Building Time Series Dataset and, to reduce annotation costs, propose generating pseudo time series from single-year maps by simulating common temporal transformations. Experimental results show that MapSAM2 learns temporal associations effectively and can accurately segment and link buildings in time series under limited supervision or using pseudo videos. We will release both our dataset and code to support future research.

cs.CV↗

arXiv2025

Engineering 2D square lattice Hubbard models in 90$^\circ$ twisted Ge/SnX (X=S, Se) moiré supperlattices

Due to the large-period superlattices emerging in moire two-dimensional (2D) materials, electronic states in such systems exhibit low energy flat bands that can be used to simulate strongly correlated physics in a highly tunable setup. While many investigations have thus far focused on moire flat bands and emergent correlated electron physics in triangular, honeycomb and quasi-one-dimensional lattices, tunable moire realizations of square lattices subject to strong correlations remain elusive. Here we propose a feasible scheme to construct moire square lattice systems by twisting two or more layers of 2D materials in a rectangular lattice by 90 degrees. We demonstrate the concept with twisted GeX/SnX (X=S, Se) moire superlattices and calculate their electronic structures from first principles. We show that the lowest conduction flat band in these systems can be described by a square lattice Hubbard model with parameters which can be controlled by varying the choice of host materials, number of layers, and external electric fields. In particular, twisted double bilayer GeSe realizes a square lattice Hubbard model with strong frustration due to the next nearest neighbour hopping that could host unconventional superconductivity{, in close analogy to the Hubbard model for copper-oxygen planes of cuprate high-temperature superconductors}. The presented scheme uses 90-degree twisted 2D materials with rectangular unit cells as a promising platform for realizing the physical phenomena of square lattice Hubbard models, establishing a new route for studying its rich phase diagram of magnetism, charge order, and unconventional superconductivity in a highly tunable setting.

cond-mat.str-el↗

arXiv2025

Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model

We present Ring-1T, the first open-source, state-of-the-art thinking model with a trillion-scale parameter. It features 1 trillion total parameters and activates approximately 50 billion per token. Training such models at a trillion-parameter scale introduces unprecedented challenges, including train-inference misalignment, inefficiencies in rollout processing, and bottlenecks in the RL system. To address these, we pioneer three interconnected innovations: (1) IcePop stabilizes RL training via token-level discrepancy masking and clipping, resolving instability from training-inference mismatches; (2) C3PO++ improves resource utilization for long rollouts under a token budget by dynamically partitioning them, thereby obtaining high time efficiency; and (3) ASystem, a high-performance RL framework designed to overcome the systemic bottlenecks that impede trillion-parameter model training. Ring-1T delivers breakthrough results across critical benchmarks: 93.4 on AIME-2025, 86.72 on HMMT-2025, 2088 on CodeForces, and 55.94 on ARC-AGI-1. Notably, it attains a silver medal-level result on the IMO-2025, underscoring its exceptional reasoning capabilities. By releasing the complete 1T parameter MoE model to the community, we provide the research community with direct access to cutting-edge reasoning capabilities. This contribution marks a significant milestone in democratizing large-scale reasoning intelligence and establishes a new baseline for open-source model performance.

cs.CL↗