arXiv subjects

Bin Wang

Publications and source records attributed to Bin Wang.

At least 271 records · Page 15Linked to original sources

arXiv2025

Non-collective Calibrating Strategy for Time Series Forecasting

Deep learning-based approaches have demonstrated significant advancements in time series forecasting. Despite these ongoing developments, the complex dynamics of time series make it challenging to establish the rule of thumb for designing the golden model architecture. In this study, we argue that refining existing advanced models through a universal calibrating strategy can deliver substantial benefits with minimal resource costs, as opposed to elaborating and training a new model from scratch. We first identify a multi-target learning conflict in the calibrating process, which arises when optimizing variables across time steps, leading to the underutilization of the model's learning capabilities. To address this issue, we propose an innovative calibrating strategy called Socket+Plug (SoP). This approach retains an exclusive optimizer and early-stopping monitor for each predicted target within each Plug while keeping the fully trained Socket backbone frozen. The model-agnostic nature of SoP allows it to directly calibrate the performance of any trained deep forecasting models, regardless of their specific architectures. Extensive experiments on various time series benchmarks and a spatio-temporal meteorological ERA5 dataset demonstrate the effectiveness of SoP, achieving up to a 22% improvement even when employing a simple MLP as the Plug (highlighted in Figure 1). Code is available at https://github.com/hanyuki23/SoP.

cs.LG↗

arXiv2025

Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs

Large language models (LLMs) have rapidly progressed into general-purpose agents capable of solving a broad spectrum of tasks. However, current models remain inefficient at reasoning: they apply fixed inference-time compute regardless of task complexity, often overthinking simple problems while underthinking hard ones. This survey presents a comprehensive review of efficient test-time compute (TTC) strategies, which aim to improve the computational efficiency of LLM reasoning. We introduce a two-tiered taxonomy that distinguishes between L1-controllability, methods that operate under fixed compute budgets, and L2-adaptiveness, methods that dynamically scale inference based on input difficulty or model confidence. We benchmark leading proprietary LLMs across diverse datasets, highlighting critical trade-offs between reasoning performance and token usage. Compared to prior surveys on efficient reasoning, our review emphasizes the practical control, adaptability, and scalability of TTC methods. Finally, we discuss emerging trends such as hybrid thinking models and identify key challenges for future work towards making LLMs more computationally efficient, robust, and responsive to user constraints.

cs.AI↗

arXiv2025

Further study of the maximally symmetry breaking patterns in an ${\rm SU}(8)$ theory

An ${\rm SU}(8)$ theory was previously found to be the minimal simple gauge group where all three-generational Standard Model (SM) fermions can be nontrivially embedded. It is maximally broken into a subgroup of ${\rm SU}(8)\to {\cal G}_{441}\equiv {\rm SU}(4)_s \otimes {\rm SU}(4)_W \otimes {\rm U}(1)_{X_0}$ at the grand unified theory scale by the ${\rm SU}(8)$ adjoint Higgs field of $\mathbf{63_H}$. Gauge symmetries in the strong and the weak sectors are extended by one and two ranks, respectively. The sequential strong-weak-weak (SWW) symmetry breaking stages were found to generate the observed hierarchical SM quark/lepton masses as well as the Cabibbo-Kobayashi-Maskawa mixing pattern with the precise flavor identifications~[17, 20]. We further study the possible weak-strong-weak and weak-weak-strong symmetry breaking patterns, and compare with the results that we have obtained by following the SWW sequence. The two-loop renormalization group equations following both patterns are analyzed, where we cannot achieve the gauge coupling unification in the field theory framework. Through these analyses, we suggest the gauge coupling unification to be interpreted in the context of the affine Lie algebra.

hep-ph↗

arXiv2025

Detectability of the 21 cm signal with BINGO through cross-correlation with photometric surveys

21 cm intensity mapping (HI IM) can efficiently map large cosmic volumes with good redshift resolution, but systematics and foreground contamination pose major challenges for extracting accurate cosmological information. Cross-correlation with galaxy surveys offers an efficient mitigation strategy, as both datasets have largely uncorrelated systematics. We evaluate the detectability of the 21 cm signal from the BINGO radio telescope by cross-correlating with the LSST photometric survey, given their strong overlap in area and redshift. Using lognormal simulations, we model the cosmological signal in the BINGO frequency range (980 - 1260 MHz), incorporating thermal noise, foregrounds, and cleanup. The LSST simulations include uncertainties in photometric redshift (photo-z) and galaxy number density in the first three redshift intervals (mean redshift approximately equal to 0.25, 0.35, 0.45), corresponding to the expected performance after 10 years of the survey. We show that photo-z errors significantly increase the noise in the cross-correlation, reducing its statistical significance to levels comparable to those of the autocorrelation. Still, the HI signal remains detectable through the cross-correlation, even with photo-z uncertainties similar to those of the LSST. Our results corroborate the feasibility of this approach under realistic conditions and motivate further refinement of the current analysis methods.

astro-ph.CO↗

arXiv2025

Cosmological Remapping for Efficient Generation of 21 cm Intensity Mapping Mocks

We present a novel application of cosmological rescaling, or "remapping," to generate 21 cm intensity mapping mocks for different cosmologies. The remapping method allows for computationally efficient generation of N-body catalogs by rescaling existing simulations. In this work, we employ the remapping method to construct dark matter halo catalogs, starting from the Horizon Run 4 simulation with WMAP5 cosmology, and apply it to different target cosmologies, including WMAP7, Planck18 and Chevallier-Polarski-Linder (CPL) models. These catalogs are then used to simulate 21 cm intensity maps. We use the halo occupation distribution (HOD) method to populate halos with neutral hydrogen (HI) and derive 21 cm brightness temperature maps. Our results demonstrate the effectiveness of the remapping approach in generating cosmological simulations for large-scale structure studies, offering an alternative for testing observational data pipelines and performing cosmological parameter forecasts without the need for computationally expensive full N-body simulations. We also analyze the precision and limitations of the remapping, in light of the rescaling parameters $s$ and $s_m$, as well as the effects of the halo mass and box size thresholds.

astro-ph.CO↗

arXiv2025

Wavelet-based Global Orientation and Surface Reconstruction for Point Clouds

Unoriented surface reconstruction is an important task in computer graphics and has extensive applications. Based on the compact support of wavelet and orthogonality properties, classic wavelet surface reconstruction achieves good and fast reconstruction. However, this method can only handle oriented points. Despite some improved attempts for unoriented points, such as iWSR, these methods perform poorly on sparse point clouds. To address these shortcomings, we propose a wavelet-based method to represent the mollified indicator function and complete both the orientation and surface reconstruction tasks. We use the modifying kernel function to smoothen out discontinuities on the surface, aligning with the continuity of the wavelet basis function. During the calculation of coefficient, we fully utilize the properties of the convolutional kernel function to shift the modifying computation onto wavelet basis to accelerate. In addition, we propose a novel method for constructing the divergence-free function field and using them to construct the additional homogeneous constraints to improve the effectiveness and stability. Extensive experiments demonstrate that our method achieves state-of-the-art performance in both orientation and reconstruction for sparse models. We align the matrix construction with the compact support property of wavelet basis functions to further accelerate our method, resulting in efficient performance on CPU. Our source codes will be released on GitHub.

cs.CG↗

arXiv2025

Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model

Large Audio-Language Models (LALMs) have significantly advanced intelligent human-computer interaction, yet their reliance on text-based outputs limits their ability to generate natural speech responses directly, hindering seamless audio interactions. To address this, we introduce Step-Audio-AQAA, a fully end-to-end LALM designed for Audio Query-Audio Answer (AQAA) tasks. The model integrates a dual-codebook audio tokenizer for linguistic and semantic feature extraction, a 130-billion-parameter backbone LLM and a neural vocoder for high-fidelity speech synthesis. Our post-training approach employs interleaved token-output of text and audio to enhance semantic coherence and combines Direct Preference Optimization (DPO) with model merge to improve performance. Evaluations on the StepEval-Audio-360 benchmark demonstrate that Step-Audio-AQAA excels especially in speech control, outperforming the state-of-art LALMs in key areas. This work contributes a promising solution for end-to-end LALMs and highlights the critical role of token-based vocoder in enhancing overall performance for AQAA tasks.

cs.SD↗

arXiv2025

TDS-CLIP: Temporal Difference Side Network for Efficient VideoAction Recognition

Recently, large-scale pre-trained vision-language models (e.g., CLIP), have garnered significant attention thanks to their powerful representative capabilities. This inspires researchers in transferring the knowledge from these large pre-trained models to other task-specific models, e.g., Video Action Recognition (VAR) models, via particularly leveraging side networks to enhance the efficiency of parameter-efficient fine-tuning (PEFT). However, current transferring approaches in VAR tend to directly transfer the frozen knowledge from large pre-trained models to action recognition networks with minimal cost, instead of exploiting the temporal modeling capabilities of the action recognition models themselves. Therefore, in this paper, we propose a novel memory-efficient Temporal Difference Side Network (TDS-CLIP) to balance knowledge transferring and temporal modeling, avoiding backpropagation in frozen parameter models. Specifically, we introduce a Temporal Difference Adapter (TD-Adapter), which can effectively capture local temporal differences in motion features to strengthen the model's global temporal modeling capabilities. Furthermore, we designed a Side Motion Enhancement Adapter (SME-Adapter) to guide the proposed side network in efficiently learning the rich motion information in videos, thereby improving the side network's ability to capture and learn motion information. Extensive experiments are conducted on three benchmark datasets, including Something-Something V1&V2, and Kinetics-400. Experimental results show that our method achieves competitive performance in video action recognition tasks.

cs.CV↗

arXiv2025

Large Language Models Struggle with Unreasonability in Math Problems

Large Language Models (LLMs) have shown remarkable success on a wide range of math and reasoning benchmarks. However, we observe that they often struggle when faced with unreasonable math problems. Instead of recognizing these issues, models frequently proceed as if the problem is well-posed, producing incorrect answers or falling into overthinking and verbose self-correction. To systematically investigate this overlooked vulnerability, we propose the \textbf{Unreasonable Math Problems (UMP)} benchmark, designed to evaluate LLMs' ability to detect and respond to unreasonable math problem statements. Based on extensive experiments covering 19 LLMs, we find that even state-of-the-art general models like GPT-4o achieve only a score of 0.6 on UMP. While reasoning models such as DeepSeek-R1 demonstrate a higher sensitivity to unreasonable inputs, this often comes at the cost of generating overly long and meaningless responses that fail to converge. We further explore prompting and fine-tuning methods, which offer partial improvements but also introduce trade-offs, shedding light on both the potential and limitations of LLMs in this challenging setting.

cs.CL↗

arXiv2025

Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition

This work presents Pangu Embedded, an efficient Large Language Model (LLM) reasoner developed on Ascend Neural Processing Units (NPUs), featuring flexible fast and slow thinking capabilities. Pangu Embedded addresses the significant computational costs and inference latency challenges prevalent in existing reasoning-optimized LLMs. We propose a two-stage training framework for its construction. In Stage 1, the model is finetuned via an iterative distillation process, incorporating inter-iteration model merging to effectively aggregate complementary knowledge. This is followed by reinforcement learning on Ascend clusters, optimized by a latency-tolerant scheduler that combines stale synchronous parallelism with prioritized data queues. The RL process is guided by a Multi-source Adaptive Reward System (MARS), which generates dynamic, task-specific reward signals using deterministic metrics and lightweight LLM evaluators for mathematics, coding, and general problem-solving tasks. Stage 2 introduces a dual-system framework, endowing Pangu Embedded with a "fast" mode for routine queries and a deeper "slow" mode for complex inference. This framework offers both manual mode switching for user control and an automatic, complexity-aware mode selection mechanism that dynamically allocates computational resources to balance latency and reasoning depth. Experimental results on benchmarks including AIME 2024, GPQA, and LiveCodeBench demonstrate that Pangu Embedded with 7B parameters, outperforms similar-size models like Qwen3-8B and GLM4-9B. It delivers rapid responses and state-of-the-art reasoning quality within a single, unified model architecture, highlighting a promising direction for developing powerful yet practically deployable LLM reasoners.

cs.CL↗

arXiv2025

HMAD: Advancing E2E Driving with Anchored Offset Proposals and Simulation-Supervised Multi-target Scoring

End-to-end autonomous driving faces persistent challenges in both generating diverse, rule-compliant trajectories and robustly selecting the optimal path from these options via learned, multi-faceted evaluation. To address these challenges, we introduce HMAD, a framework integrating a distinctive Bird's-Eye-View (BEV) based trajectory proposal mechanism with learned multi-criteria scoring. HMAD leverages BEVFormer and employs learnable anchored queries, initialized from a trajectory dictionary and refined via iterative offset decoding (inspired by DiffusionDrive), to produce numerous diverse and stable candidate trajectories. A key innovation, our simulation-supervised scorer module, then evaluates these proposals against critical metrics including no at-fault collisions, drivable area compliance, comfortableness, and overall driving quality (i.e., extended PDM score). Demonstrating its efficacy, HMAD achieves a 44.5% driving score on the CVPR 2025 private test set. This work highlights the benefits of effectively decoupling robust trajectory generation from comprehensive, safety-aware learned scoring for advanced autonomous driving.

cs.CV↗

arXiv2025

Unconventional Hall Effect in Gapless Superconductors: Transverse Supercurrent Converted from Normal Current

A normal metallic system proximitized by a superconductor can exhibit a gapless superconducting state characterized by segmented Fermi surfaces, as confirmed experimentally. In such a state, quasiparticle states remain gapless along one direction, while a superconducting gap opens in the perpendicular direction. This anisotropy enables a novel Hall effect in gapless superconductors, termed the superconducting Hall effect (ScHE), where a longitudinal normal current carried by quasiparticles is converted into a dissipationless transverse supercurrent. Employing both the thermodynamic approach for bulk systems and quantum transport theory for a four-probe setup, we demonstrate the existence of this effect and reveal its intrinsic origin as the quasiparticle Berry curvature. The predicted ScHE can be experimentally verified via the standard angular-dependent Hall measurements performed on gapless superconductors.

cond-mat.mes-hall↗

arXiv2025

Dimension-Reduction Attack! Video Generative Models are Experts on Controllable Image Synthesis

Video generative models can be regarded as world simulators due to their ability to capture dynamic, continuous changes inherent in real-world environments. These models integrate high-dimensional information across visual, temporal, spatial, and causal dimensions, enabling predictions of subjects in various status. A natural and valuable research direction is to explore whether a fully trained video generative model in high-dimensional space can effectively support lower-dimensional tasks such as controllable image generation. In this work, we propose a paradigm for video-to-image knowledge compression and task adaptation, termed \textit{Dimension-Reduction Attack} (\texttt{DRA-Ctrl}), which utilizes the strengths of video models, including long-range context modeling and flatten full-attention, to perform various generation tasks. Specially, to address the challenging gap between continuous video frames and discrete image generation, we introduce a mixup-based transition strategy that ensures smooth adaptation. Moreover, we redesign the attention structure with a tailored masking mechanism to better align text prompts with image-level control. Experiments across diverse image generation tasks, such as subject-driven and spatially conditioned generation, show that repurposed video models outperform those trained directly on images. These results highlight the untapped potential of large-scale video generators for broader visual applications. \texttt{DRA-Ctrl} provides new insights into reusing resource-intensive video models and lays foundation for future unified generative models across visual modalities. The project page is https://dra-ctrl-2025.github.io/DRA-Ctrl/.

cs.CV↗

arXiv2025

Scaling-up Perceptual Video Quality Assessment

The data scaling law has been shown to significantly enhance the performance of large multi-modal models (LMMs) across various downstream tasks. However, in the domain of perceptual video quality assessment (VQA), the potential of scaling law remains unprecedented due to the scarcity of labeled resources and the insufficient scale of datasets. To address this, we propose \textbf{OmniVQA}, an efficient framework designed to efficiently build high-quality, human-in-the-loop VQA multi-modal instruction databases (MIDBs). We then scale up to create \textbf{OmniVQA-Chat-400K}, the largest MIDB in the VQA field concurrently. Our focus is on the technical and aesthetic quality dimensions, with abundant in-context instruction data to provide fine-grained VQA knowledge. Additionally, we have built the \textbf{OmniVQA-MOS-20K} dataset to enhance the model's quantitative quality rating capabilities. We then introduce a \textbf{complementary} training strategy that effectively leverages the knowledge from datasets for quality understanding and quality rating tasks. Furthermore, we propose the \textbf{OmniVQA-FG (fine-grain)-Benchmark} to evaluate the fine-grained performance of the models. Our results demonstrate that our models achieve state-of-the-art performance in both quality understanding and rating tasks.

cs.CV↗

arXiv2025

TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization

The Key-Value (KV) cache in generative large language models (LLMs) introduces substantial memory overhead. Existing works mitigate this burden by offloading or compressing the KV cache. However, loading the entire cache incurs significant latency due to PCIe bandwidth bottlenecks in CPU-GPU communication, while aggressive compression causes notable performance degradation. We identify that certain layers in the LLM need to maintain global information and are unsuitable for selective loading. In contrast, other layers primarily focus on a few tokens with dominant activations that potentially incur substantial quantization error. This observation leads to a key insight that loading dominant tokens and quantizing all tokens can complement each other. Building on this insight, we propose a hybrid compression method, TailorKV, which seamlessly integrates quantization and offloading. TailorKV develops an inference framework along with a hardware-friendly implementation that leverages these complementary characteristics. Extensive long-context evaluations exhibit that TailorKV achieves nearly lossless performance under aggressive compression settings, outperforming the state-of-the-art. Particularly, the Llama-3.1-8B with 128k context can be served within a single RTX 3090 GPU, reaching 82 ms per token during decoding.

cs.CL↗

arXiv2025

TimeCF: A TimeMixer-Based Model with adaptive Convolution and Sharpness-Aware Minimization Frequency Domain Loss for long-term time seris forecasting

Recent studies have shown that by introducing prior knowledge, multi-scale analysis of complex and non-stationary time series in real environments can achieve good results in the field of long-term forecasting. However, affected by channel-independent methods, models based on multi-scale analysis may produce suboptimal prediction results due to the autocorrelation between time series labels, which in turn affects the generalization ability of the model. To address this challenge, we are inspired by the idea of sharpness-aware minimization and the recently proposed FreDF method and design a deep learning model TimeCF for long-term time series forecasting based on the TimeMixer, combined with our designed adaptive convolution information aggregation module and Sharpness-Aware Minimization Frequency Domain Loss (SAMFre). Specifically, TimeCF first decomposes the original time series into sequences of different scales. Next, the same-sized convolution modules are used to adaptively aggregate information of different scales on sequences of different scales. Then, decomposing each sequence into season and trend parts and the two parts are mixed at different scales through bottom-up and top-down methods respectively. Finally, different scales are aggregated through a Feed-Forward Network. What's more, extensive experimental results on different real-world datasets show that our proposed TimeCF has excellent performance in the field of long-term forecasting.

cs.LG↗

arXiv2025

FG-CLIP: Fine-Grained Visual and Textual Alignment

Contrastive Language-Image Pre-training (CLIP) excels in multimodal tasks such as image-text retrieval and zero-shot classification but struggles with fine-grained understanding due to its focus on coarse-grained short captions. To address this, we propose Fine-Grained CLIP (FG-CLIP), which enhances fine-grained understanding through three key innovations. First, we leverage large multimodal models to generate 1.6 billion long caption-image pairs for capturing global-level semantic details. Second, a high-quality dataset is constructed with 12 million images and 40 million region-specific bounding boxes aligned with detailed captions to ensure precise, context-rich representations. Third, 10 million hard fine-grained negative samples are incorporated to improve the model's ability to distinguish subtle semantic differences. We construct a comprehensive dataset, termed FineHARD, by integrating high-quality region-specific annotations with hard fine-grained negative samples. Corresponding training methods are meticulously designed for these data. Extensive experiments demonstrate that FG-CLIP outperforms the original CLIP and other state-of-the-art methods across various downstream tasks, including fine-grained understanding, open-vocabulary object detection, image-text retrieval, and general multimodal benchmarks. These results highlight FG-CLIP's effectiveness in capturing fine-grained image details and improving overall model performance. The data, code, and models are available at https://github.com/360CVGroup/FG-CLIP.

cs.CV↗

arXiv2025

Accelerating Diffusion-based Super-Resolution with Dynamic Time-Spatial Sampling

Diffusion models have gained attention for their success in modeling complex distributions, achieving impressive perceptual quality in SR tasks. However, existing diffusion-based SR methods often suffer from high computational costs, requiring numerous iterative steps for training and inference. Existing acceleration techniques, such as distillation and solver optimization, are generally task-agnostic and do not fully leverage the specific characteristics of low-level tasks like super-resolution (SR). In this study, we analyze the frequency- and spatial-domain properties of diffusion-based SR methods, revealing key insights into the temporal and spatial dependencies of high-frequency signal recovery. Specifically, high-frequency details benefit from concentrated optimization during early and late diffusion iterations, while spatially textured regions demand adaptive denoising strategies. Building on these observations, we propose the Time-Spatial-aware Sampling strategy (TSS) for the acceleration of Diffusion SR without any extra training cost. TSS combines Time Dynamic Sampling (TDS), which allocates more iterations to refining textures, and Spatial Dynamic Sampling (SDS), which dynamically adjusts strategies based on image content. Extensive evaluations across multiple benchmarks demonstrate that TSS achieves state-of-the-art (SOTA) performance with significantly fewer iterations, improving MUSIQ scores by 0.2 - 3.0 and outperforming the current acceleration methods with only half the number of steps.

cs.CV↗