Search

Large language models (LLMs) have demonstrated remarkable proficiency in mainstream academic disciplines such as mathematics, physics, and computer science. However, human knowledge encompasses over 200 specialized disciplines, far exceeding the scope of existing benchmarks. The capabilities of LLMs in many of these specialized fields-particularly in light industry, agriculture, and service-oriented disciplines-remain inadequately evaluated. To address this gap, we present SuperGPQA, a comprehensive benchmark that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines. Our benchmark employs a novel Human-LLM collaborative filtering mechanism to eliminate trivial or ambiguous questions through iterative refinement based on both LLM responses and expert feedback. Our experimental results reveal significant room for improvement in the performance of current state-of-the-art LLMs across diverse knowledge domains (e.g., the reasoning-focused model DeepSeek-R1 achieved the highest accuracy of 61.82% on SuperGPQA), highlighting the considerable gap between current model capabilities and artificial general intelligence. Additionally, we present comprehensive insights from our management of a large-scale annotation process, involving over 80 expert annotators and an interactive Human-LLM collaborative system, offering valuable methodological guidance for future research initiatives of comparable scope.

Pattern search is crucial in numerous analytic applications for retrieving data entries akin to the query. Content Addressable Memories (CAMs), an in-memory computing fabric, directly compare input queries with stored entries through embedded comparison logic, facilitating fast parallel pattern search in memory. While conventional CAM designs offer exact match functionality, they are inadequate for meeting the approximate search needs of emerging data-intensive applications. Some recent CAM designs propose approximate matching functions, but they face limitations such as excessively large cell area or the inability to precisely control the degree of approximation. In this paper, we propose TAP-CAM, a novel ferroelectric field effect transistor (FeFET) based ternary CAM (TCAM) capable of both exact and tunable approximate matching. TAP-CAM employs a compact 2FeFET-2R cell structure as the entry storage unit, and similarities in Hamming distances between input queries and stored entries are measured using an evaluation transistor associated with the matchline of CAM array. The operation, robustness and performance of the proposed design at array level have been discussed and evaluated, respectively. We conduct a case study of K-nearest neighbor (KNN) search to benchmark the proposed TAP-CAM at application level. Results demonstrate that compared to 16T CMOS CAM with exact match functionality, TAP-CAM achieves a 16.95x energy improvement, along with a 3.06% accuracy enhancement. Compared to 2FeFET TCAM with approximate match functionality, TAP-CAM achieves a 6.78x energy improvement.

Knowledge Tracing (KT) is crucial in education assessment, which focuses on depicting students' learning states and assessing students' mastery of subjects. With the rise of modern online learning platforms, particularly massive open online courses (MOOCs), an abundance of interaction data has greatly advanced the development of the KT technology. Previous research commonly adopts deterministic representation to capture students' knowledge states, which neglects the uncertainty during student interactions and thus fails to model the true knowledge state in learning process. In light of this, we propose an Uncertainty-Aware Knowledge Tracing model (UKT) which employs stochastic distribution embeddings to represent the uncertainty in student interactions, with a Wasserstein self-attention mechanism designed to capture the transition of state distribution in student learning behaviors. Additionally, we introduce the aleatory uncertainty-aware contrastive learning loss, which strengthens the model's robustness towards different types of uncertainties. Extensive experiments on six real-world datasets demonstrate that UKT not only significantly surpasses existing deep learning-based models in KT prediction, but also shows unique advantages in handling the uncertainty of student interactions., Comment: Accepted by AAAI 2025

We present a systematic investigation of extremely X-ray variable active galactic nuclei (AGNs) in the $\approx 5.3~{\rm deg}^2$ XMM-SERVS XMM-LSS region. Eight variable AGNs are identified with rest-frame 2 keV flux density variability amplitudes around 6-12. We comprehensively analyze the X-ray and multiwavelength data to probe the origin of their extreme X-ray variability. It is found that their extreme X-ray variability can be ascribed to changing accretion state or changing obscuration from dust-free absorbers. For five AGNs, their X-ray variability is attributed to changing accretion state, supported by contemporaneous multiwavelength variability and the absence of X-ray absorption in the low-state spectra. With new Multiple Mirror Telescope (MMT) spectra for four of these sources, we confirm one changing-look AGN. One MMT AGN lacks multi-epoch spectroscopic observations, while the other two AGNs do not exhibit changing-look behavior, likely because the MMT observations did not capture their high states. The X-ray variability of the other three AGNs is explained by changing obscuration, and they show only mild long-term optical/IR variability. The absorbers of these sources are likely clumpy accretion-disk winds, with variable column densities and covering factors along the lines of sight., Comment: 32 pages, 13 figures, 5 tables, accepted for publication in the Astrophysical Journal

The emergence of long-context large language models (LLMs) offers a promising alternative to traditional retrieval-augmented generation (RAG) for processing extensive documents. However, the computational overhead of long-context inference, particularly in managing key-value (KV) caches, presents significant efficiency challenges. While Speculative Decoding (SD) traditionally accelerates inference using smaller draft models, its effectiveness diminishes substantially in long-context scenarios due to memory-bound KV cache operations. We present Retrieval-Augmented Speculative Decoding (RAPID), which leverages RAG for both accelerating and enhancing generation quality in long-context inference. RAPID introduces the RAG drafter-a draft LLM operating on shortened retrieval contexts-to speculate on the generation of long-context target LLMs. Our approach enables a new paradigm where same-scale or even larger LLMs can serve as RAG drafters while maintaining computational efficiency. To fully leverage the potentially superior capabilities from stronger RAG drafters, we develop an inference-time knowledge transfer dynamic that enriches the target distribution by RAG. Extensive experiments on the LLaMA-3.1 and Qwen2.5 backbones demonstrate that RAPID effectively integrates the strengths of both approaches, achieving significant performance improvements (e.g., from 39.33 to 42.83 on InfiniteBench for LLaMA-3.1-8B) with more than 2x speedups. Our analyses reveal that RAPID achieves robust acceleration beyond 32K context length and demonstrates superior generation quality in real-world applications.

We investigate the Cauchy problem for a fluid-particle interaction model in $\mathbb{R}^3$. This model consists of the compressible barotropic Navier-Stokes equations and the Vlasov-Fokker-Planck equation coupled together via the density-dependent friction force. Due to the strong coupling caused by the friction force, it is a challenging problem to construct the global existence and optimal decay rates of strong solutions. In this paper, by assuming that the $H^2$-norm of the initial data is sufficiently small, we establish the global well-posedness of strong solutions. Furthermore, if the $L^1$-norm of initial data is bounded, then we achieve the optimal decay rates of strong solutions and their gradients in $L^2$-norm. The proofs rely on developing refined energy estimates and exploiting the frequency decomposition method. In addition, for the periodic domain case, our global strong solutions decay exponentially., Comment: 38pages

Building articulated objects is a key challenge in computer vision. Existing methods often fail to effectively integrate information across different object states, limiting the accuracy of part-mesh reconstruction and part dynamics modeling, particularly for complex multi-part articulated objects. We introduce ArtGS, a novel approach that leverages 3D Gaussians as a flexible and efficient representation to address these issues. Our method incorporates canonical Gaussians with coarse-to-fine initialization and updates for aligning articulated part information across different object states, and employs a skinning-inspired part dynamics modeling module to improve both part-mesh reconstruction and articulation learning. Extensive experiments on both synthetic and real-world datasets, including a new benchmark for complex multi-part objects, demonstrate that ArtGS achieves state-of-the-art performance in joint parameter estimation and part mesh reconstruction. Our approach significantly improves reconstruction quality and efficiency, especially for multi-part articulated objects. Additionally, we provide comprehensive analyses of our design choices, validating the effectiveness of each component to highlight potential areas for future improvement.

Biophysical processes within living systems rely on encounters and interactions between molecules in complex environments such as cells. They are often described by anomalous diffusion transport. Recent advances in single-molecule microscopy and particle-tracking techniques have yielded an abundance of data in the form of videos and trajectories that contain critical information about these biologically significant processes. However, standard approaches for characterizing anomalous diffusion from these measurements often struggle in cases of practical interest, e.g. due to short, noisy trajectories. Fully exploiting this data therefore requires the development of advanced analysis methods -- a core goal at the heart of the recent international Anomalous Diffusion Challenges. Here, we introduce a novel machine-learning framework, U-net 3+ for Anomalous Diffusion analysis enhanced with Mixture Estimates (U-AnD-ME), that applies a U-Net 3+ based neural network alongside Gaussian mixture models to enable highly accurate characterisation of single-particle tracking data. In the 2024 Anomalous Diffusion Challenge, U-AnD-ME outperformed all other participating methods for the analysis of two-dimensional anomalous diffusion trajectories at both single-trajectory and ensemble levels. Using a large dataset inspired by the Challenge, we further characterize the performance of U-AnD-ME in segmenting trajectories and inferring anomalous diffusion properties.

Coronal mass ejections (CMEs) are the eruptions of magnetised plasma from the Sun and are considered the main driver of adverse space weather events. Hence, undrstanding its formation process, particularly the magnetic topology, is critical for accurate space weather prediction. Here, based on imaging observations and three-dimensional (3D) data-constrained thermodynamic magnetohydrodynamical (MHD) simulation in spherical coordinates, we exhibit the birth of a CME with intricate magnetic structure from multiple active regions (ARs) due to 3D magnetic reconnection. It is observed as a coronal jet between active regions, accompanied by the back-flowing of filament materials along the jet spine after the passage of the eruptive filament. This jet connects two dimming regions within different active regions. This is an observational proxy of 3D magnetic reconnection between the CME flux rope and the null-point magnetic field lines crossing active regions. Hereafter, the thermodynamic data-constrained MHD simulation successfully reproduces the observed jet and the reconnection process that flux ropes partake in, leading to a CME flux rope with a complex magnetic structure distinct from its progenitor. The generality of this scenario is then validated by data-inspired MHD simulations in a simple multipolar magnetic configuration. This work demonstrates the role of multiple active regions in forming CMEs with intricate magnetic structures. On the one hand, a non-coherent flux rope where not all twisted magnetic field lines wind around one common axis is naturally formed. On the other hand, our findings suggest that the topology of a real CME flux rope may not be solely determined by a single active region, particularly during periods of solar maximum., Comment: 19 pages, 8 figures, accepted for publication in ApJ

Human drivers adeptly navigate complex scenarios by utilizing rich attentional semantics, but the current autonomous systems struggle to replicate this ability, as they often lose critical semantic information when converting 2D observations into 3D space. In this sense, it hinders their effective deployment in dynamic and complex environments. Leveraging the superior scene understanding and reasoning abilities of Vision-Language Models (VLMs), we propose VLM-E2E, a novel framework that uses the VLMs to enhance training by providing attentional cues. Our method integrates textual representations into Bird's-Eye-View (BEV) features for semantic supervision, which enables the model to learn richer feature representations that explicitly capture the driver's attentional semantics. By focusing on attentional semantics, VLM-E2E better aligns with human-like driving behavior, which is critical for navigating dynamic and complex environments. Furthermore, we introduce a BEV-Text learnable weighted fusion strategy to address the issue of modality importance imbalance in fusing multimodal information. This approach dynamically balances the contributions of BEV and text features, ensuring that the complementary information from visual and textual modality is effectively utilized. By explicitly addressing the imbalance in multimodal fusion, our method facilitates a more holistic and robust representation of driving environments. We evaluate VLM-E2E on the nuScenes dataset and demonstrate its superiority over state-of-the-art approaches, showcasing significant improvements in performance.

We report on a search for weakly interacting massive particle (WIMP) dark matter (DM) via elastic DM-xenon-nucleus interactions in the XENONnT experiment. We combine datasets from the first and second science campaigns resulting in a total exposure of $3.1\;\text{tonne}\times\text{year}$. In a blind analysis of nuclear recoil events with energies above $3.8\,\mathrm{keV_{NR}}$, we find no significant excess above background. We set new upper limits on the spin-independent WIMP-nucleon scattering cross-section for WIMP masses above $10\,\mathrm{GeV}/c^2$ with a minimum of $1.7\,\times\,10^{-47}\,\mathrm{cm^2}$ at $90\,\%$ confidence level for a WIMP mass of $30\,\mathrm{GeV}/c^2$. We achieve a best median sensitivity of $1.4\,\times\,10^{-47}\,\mathrm{cm^2}$ for a $41\,\mathrm{GeV}/c^2$ WIMP. Compared to the result from the first XENONnT science dataset, we improve our sensitivity by a factor of up to 1.8., Comment: Limits are included in the submission file

Recent efforts have extended textual LLMs to the speech domain. Yet, a key challenge remains, which is balancing speech understanding and generation while avoiding catastrophic forgetting when integrating acoustically rich codec-based representations into models originally trained on text. In this work, we propose a novel approach that leverages continual pre-training (CPT) on a pre-trained textual LLM to create a codec-based speech language model. This strategy mitigates the modality gap between text and speech, preserving the linguistic reasoning of the original model while enabling high-fidelity speech synthesis. We validate our approach with extensive experiments across multiple tasks, including automatic speech recognition, text-to-speech, speech-to-text translation, and speech-to-speech translation (S2ST), demonstrating that our model achieves superior TTS performance and, notably, the first end-to-end S2ST system based on neural codecs.

To alleviate memory burden during inference of large language models (LLMs), numerous studies have focused on compressing the KV cache by exploring aspects such as attention sparsity. However, these techniques often require a pre-defined cache budget; as the optimal budget varies with different input lengths and task types, it limits their practical deployment accepting open-domain instructions. To address this limitation, we propose a new KV cache compression objective: to always ensure the full-cache performance regardless of specific inputs, while maximizing KV cache pruning as much as possible. To achieve this goal, we introduce a novel KV cache compression method dubbed DBudgetKV, which features an attention-based metric to signal when the remaining KV cache is unlikely to match the full-cache performance, then halting the pruning process. Empirical evaluation spanning diverse context lengths, task types, and model sizes suggests that our method achieves lossless KV pruning effectively and robustly, exceeding 25% compression ratio on average. Furthermore, our method is easy to integrate within LLM inference, not only optimizing memory space, but also showing reduced inference time compared to existing methods.

Graph data has become a pivotal modality due to its unique ability to model relational datasets. However, real-world graph data continues to grow exponentially, resulting in a quadratic increase in the complexity of most graph algorithms as graph sizes expand. Although graph condensation (GC) methods have been proposed to address these scalability issues, existing approaches often treat the training set as static, overlooking the evolving nature of real-world graph data. This limitation leads to inefficiencies when condensing growing training sets. In this paper, we introduce GECC (Graph Evolving Clustering Condensation), a scalable graph condensation method designed to handle large-scale and evolving graph data. GECC employs a traceable and efficient approach by performing class-wise clustering on aggregated features. Furthermore, it can inherits previous condensation results as clustering centroids when the condensed graph expands, thereby attaining an evolving capability. This methodology is supported by robust theoretical foundations and demonstrates superior empirical performance. Comprehensive experiments show that GECC achieves better performance than most state-of-the-art graph condensation methods while delivering an around 1,000x speedup on large datasets., Comment: 16 pages, 6 figures

Tumor is a leading cause of death worldwide, with an estimated 10 million deaths attributed to tumor-related diseases every year. AI-driven tumor recognition unlocks new possibilities for more precise and intelligent tumor screening and diagnosis. However, the progress is heavily hampered by the scarcity of annotated datasets, which demands extensive annotation efforts by radiologists. To tackle this challenge, we introduce FreeTumor, an innovative Generative AI (GAI) framework to enable large-scale tumor synthesis for mitigating data scarcity. Specifically, FreeTumor effectively leverages a combination of limited labeled data and large-scale unlabeled data for tumor synthesis training. Unleashing the power of large-scale data, FreeTumor is capable of synthesizing a large number of realistic tumors on images for augmenting training datasets. To this end, we create the largest training dataset for tumor synthesis and recognition by curating 161,310 publicly available Computed Tomography (CT) volumes from 33 sources, with only 2.3% containing annotated tumors. To validate the fidelity of synthetic tumors, we engaged 13 board-certified radiologists in a Visual Turing Test to discern between synthetic and real tumors. Rigorous clinician evaluation validates the high quality of our synthetic tumors, as they achieved only 51.1% sensitivity and 60.8% accuracy in distinguishing our synthetic tumors from real ones. Through high-quality tumor synthesis, FreeTumor scales up the recognition training datasets by over 40 times, showcasing a notable superiority over state-of-the-art AI methods including various synthesis methods and foundation models. These findings indicate promising prospects of FreeTumor in clinical applications, potentially advancing tumor treatments and improving the survival rates of patients.

Recently, denoising diffusion models have achieved promising results in 2D image generation and editing. Instruct-NeRF2NeRF (IN2N) introduces the success of diffusion into 3D scene editing through an "Iterative dataset update" (IDU) strategy. Though achieving fascinating results, IN2N suffers from problems of blurry backgrounds and trapping in local optima. The first problem is caused by IN2N's lack of efficient guidance for background maintenance, while the second stems from the interaction between image editing and NeRF training during IDU. In this work, we introduce DualNeRF to deal with these problems. We propose a dual-field representation to preserve features of the original scene and utilize them as additional guidance to the model for background maintenance during IDU. Moreover, a simulated annealing strategy is embedded into IDU to endow our model with the power of addressing local optima issues. A CLIP-based consistency indicator is used to further improve the editing quality by filtering out low-quality edits. Extensive experiments demonstrate that our method outperforms previous methods both qualitatively and quantitatively.

To evaluate end-to-end autonomous driving systems, a simulation environment based on Novel View Synthesis (NVS) techniques is essential, which synthesizes photo-realistic images and point clouds from previously recorded sequences under new vehicle poses, particularly in cross-lane scenarios. Therefore, the development of a multi-lane dataset and benchmark is necessary. While recent synthetic scene-based NVS datasets have been prepared for cross-lane benchmarking, they still lack the realism of captured images and point clouds. To further assess the performance of existing methods based on NeRF and 3DGS, we present the first multi-lane dataset registering parallel scans specifically for novel driving view synthesis dataset derived from real-world scans, comprising 25 groups of associated sequences, including 16,000 front-view images, 64,000 surround-view images, and 16,000 LiDAR frames. All frames are labeled to differentiate moving objects from static elements. Using this dataset, we evaluate the performance of existing approaches in various testing scenarios at different lanes and distances. Additionally, our method provides the solution for solving and assessing the quality of multi-sensor poses for multi-modal data alignment for curating such a dataset in real-world. We plan to continually add new sequences to test the generalization of existing methods across different scenarios. The dataset is released publicly at the project page: https://nizqleo.github.io/paralane-dataset/., Comment: Accepted by International Conference on 3D Vision (3DV) 2025

In this study, we present a comprehensive analysis of an unidentified point-like ultra-high-energy (UHE) $\gamma$-ray source, designated as 1LHAASO J1740+0948u, situated in the vicinity of the middle-aged pulsar PSR J1740+1000. The detection significance reached 17.1$\sigma$ (9.4$\sigma$) above 25$\,$TeV (100$\,$TeV). The source energy spectrum extended up to 300$\,$TeV, which was well fitted by a log-parabola function with $N0 = (1.93\pm0.23) \times 10^{-16} \rm{TeV^{-1}\,cm^{-2}\,s^{-2}}$, $\alpha = 2.14\pm0.27$, and $\beta = 1.20\pm0.41$ at E0 = 30$\,$TeV. The associated pulsar, PSR J1740+1000, resides at a high galactic latitude and powers a bow-shock pulsar wind nebula (BSPWN) with an extended X-ray tail. The best-fit position of the gamma-ray source appeared to be shifted by $0.2^{\circ}$ with respect to the pulsar position. As the (i) currently identified pulsar halos do not demonstrate such offsets, and (ii) centroid of the gamma-ray emission is approximately located at the extension of the X-ray tail, we speculate that the UHE $\gamma$-ray emission may originate from re-accelerated electron/positron pairs that are advected away in the bow-shock tail., Comment: Corrected spelling errors in several author names

Transformer-based and CNN-based methods demonstrate strong performance in long-term time series forecasting. However, their high computational and storage requirements can hinder large-scale deployment. To address this limitation, we propose integrating lightweight MLP with advanced architectures using knowledge distillation (KD). Our preliminary study reveals different models can capture complementary patterns, particularly multi-scale and multi-period patterns in the temporal and frequency domains. Based on this observation, we introduce TimeDistill, a cross-architecture KD framework that transfers these patterns from teacher models (e.g., Transformers, CNNs) to MLP. Additionally, we provide a theoretical analysis, demonstrating that our KD approach can be interpreted as a specialized form of mixup data augmentation. TimeDistill improves MLP performance by up to 18.6%, surpassing teacher models on eight datasets. It also achieves up to 7X faster inference and requires 130X fewer parameters. Furthermore, we conduct extensive evaluations to highlight the versatility and effectiveness of TimeDistill.

Air-sea interface fluxes significantly impact the reliability and efficiency of maritime communication. Compared to sparse in-situ ocean observations, satellite remote sensing data offers broader coverage and extended temporal span. This study utilizes COARE V3.5 algorithm to calculate momentum flux, sensible heat flux, and latent heat flux at the air-sea interface, based on satellite synthetic aperture radar (SAR) wind speed data, reanalysis data, and buoy measurements, combined with neural network methods. Findings indicate that SAR wind speed data corrected via neural networks show improved consistency with buoy-measured wind speeds in flux calculations. Specifically, the bias in friction velocity decreased from -0.03 m/s to 0.01 m/s, wind stress bias from -0.03 N/m^2 to 0.00 N/m^2, drag coefficient bias from -0.29 to -0.21, latent heat flux bias from -8.32 W/m^2 to 5.41 W/m^2, and sensible heat flux bias from 0.67 W/m^2 to 0.06 W/m^2. Results suggest that the neural network-corrected SAR wind speed data can provide more reliable environmental data for maritime communication., Comment: in Chinese language, Mobile Communications

Generative recommendation (GR) is an emerging paradigm where user actions are tokenized into discrete token patterns and autoregressively generated as predictions. However, existing GR models tokenize each action independently, assigning the same fixed tokens to identical actions across all sequences without considering contextual relationships. This lack of context-awareness can lead to suboptimal performance, as the same action may hold different meanings depending on its surrounding context. To address this issue, we propose ActionPiece to explicitly incorporate context when tokenizing action sequences. In ActionPiece, each action is represented as a set of item features, which serve as the initial tokens. Given the action sequence corpora, we construct the vocabulary by merging feature patterns as new tokens, based on their co-occurrence frequency both within individual sets and across adjacent sets. Considering the unordered nature of feature sets, we further introduce set permutation regularization, which produces multiple segmentations of action sequences with the same semantics. Experiments on public datasets demonstrate that ActionPiece consistently outperforms existing action tokenization methods, improving NDCG@$10$ by $6.00\%$ to $12.82\%$.

StarCraft II is a complex and dynamic real-time strategy (RTS) game environment, which is very suitable for artificial intelligence and reinforcement learning research. To address the problem of Large Language Model(LLM) learning in complex environments through self-reflection, we propose a Reflection of Episodes(ROE) framework based on expert experience and self-experience. This framework first obtains key information in the game through a keyframe selection method, then makes decisions based on expert experience and self-experience. After a game is completed, it reflects on the previous experience to obtain new self-experience. Finally, in the experiment, our method beat the robot under the Very Hard difficulty in TextStarCraft II. We analyze the data of the LLM in the process of the game in detail, verified its effectiveness.

Decades of research have produced robust methods for unsupervised data visualization, yet supervised visualization$\unicode{x2013}$where expert labels guide representations$\unicode{x2013}$remains underexplored, as most supervised approaches prioritize classification over visualization. Recently, RF-PHATE, a diffusion-based manifold learning method leveraging random forests and information geometry, marked significant progress in supervised visualization. However, its lack of an explicit mapping function limits scalability and prevents application to unseen data, posing challenges for large datasets and label-scarce scenarios. To overcome these limitations, we introduce Random Forest Autoencoders (RF-AE), a neural network-based framework for out-of-sample kernel extension that combines the flexibility of autoencoders with the supervised learning strengths of random forests and the geometry captured by RF-PHATE. RF-AE enables efficient out-of-sample supervised visualization and outperforms existing methods, including RF-PHATE's standard kernel extension, in both accuracy and interpretability. Additionally, RF-AE is robust to the choice of hyper-parameters and generalizes to any kernel-based dimensionality reduction method.

Tensor decomposition has emerged as a powerful framework for feature extraction in multi-modal biomedical data. In this review, we present a comprehensive analysis of tensor decomposition methods such as Tucker, CANDECOMP/PARAFAC, spiked tensor decomposition, etc. and their diverse applications across biomedical domains such as imaging, multi-omics, and spatial transcriptomics. To systematically investigate the literature, we applied a topic modeling-based approach that identifies and groups distinct thematic sub-areas in biomedicine where tensor decomposition has been used, thereby revealing key trends and research directions. We evaluated challenges related to the scalability of latent spaces along with obtaining the optimal rank of the tensor, which often hinder the extraction of meaningful features from increasingly large and complex datasets. Additionally, we discuss recent advances in quantum algorithms for tensor decomposition, exploring how quantum computing can be leveraged to address these challenges. Our study includes a preliminary resource estimation analysis for quantum computing platforms and examines the feasibility of implementing quantum-enhanced tensor decomposition methods on near-term quantum devices. Collectively, this review not only synthesizes current applications and challenges of tensor decomposition in biomedical analyses but also outlines promising quantum computing strategies to enhance its impact on deriving actionable insights from complex biomedical data., Comment: 31 pages, 7 figures

Estimating Mutual Information (MI), a key measure of dependence of random quantities without specific modelling assumptions, is a challenging problem in high dimensions. We propose a novel mutual information estimator based on parametrizing conditional densities using normalizing flows, a deep generative model that has gained popularity in recent years. This estimator leverages a block autoregressive structure to achieve improved bias-variance trade-offs on standard benchmark tasks., Comment: 23 pages, 17 figures

Retrieval augmented generation (RAG) holds great promise in addressing challenges associated with long video understanding. These methods retrieve useful moments from long videos for their presented tasks, thereby enabling multimodal large language models (MLLMs) to generate high-quality answers in a cost-effective way. In this work, we present MomentSeeker, a comprehensive benchmark to evaluate retrieval models' performance in handling general long-video moment retrieval (LVMR) tasks. MomentSeeker offers three key advantages. First, it incorporates long videos of over 500 seconds on average, making it the first benchmark specialized for long-video moment retrieval. Second, it covers a wide range of task categories (including Moment Search, Caption Alignment, Image-conditioned Moment Search, and Video-conditioned Moment Search) and diverse application scenarios (e.g., sports, movies, cartoons, and ego), making it a comprehensive tool for assessing retrieval models' general LVMR performance. Additionally, the evaluation tasks are carefully curated through human annotation, ensuring the reliability of assessment. We further fine-tune an MLLM-based LVMR retriever on synthetic data, which demonstrates strong performance on our benchmark. We perform extensive experiments with various popular multimodal retrievers based on our benchmark, whose results highlight the challenges of LVMR and limitations for existing methods. Our created resources will be shared with community to advance future research in this field.

In this work, we trace the complete life cycle of individual GMCs in high-resolution Milky Way-mass galaxy simulations to determine how different stellar feedback mechanisms and galactic-scale processes govern cloud lifetimes, mass evolution, and local star formation efficiency (SFE). We identify GMCs in simulated galaxies and track their evolution using cloud evolution trees. Via cloud evolution trees, we quantify the lifetimes and SFE of GMCs. We further apply our diagnostics on a suite of simulations with varying star formation and stellar feedback subgrid models and explore their impact together with galactic environments to the GMC life cycles. Our analysis reveals that GMCs undergo dynamic evolution, characterized by continuous gas accretion, gravitational collapse, and star formation, followed by disruption due to stellar feedback. The accretion process sustains the gas content throughout most of the GMC life cycles, resulting in a positive correlation between GMC lifetimes and their maximum masses. The GMC lifetimes range from a few to several tens of Myr, with two distinct dynamical modes: (1) GMCs near the galactic center experience strong tidal disturbances, prolonging their lifetimes when they remain marginally unbound; (2) those in the outer regions are less affected by tides, remain gravitationally bound, and evolve more rapidly. In all model variations, we observe that GMC-scale SFE correlates with the baryonic surface density of GMCs, consistent with previous studies of isolated GMCs. Additionally, we emphasize the critical role of galactic shear in regulating GMC-scale star formation and refine the correlation between local SFE and surface density by including its effects. These findings demonstrate how stellar feedback and galactic-scale dynamics jointly shape GMC-scale star formation in realistic galactic environments., Comment: 12 pages, 9 figures; submitted to A&A; abstract slightly abridged; comments welcome

Instruction Fine-tuning (IFT) can enhance the helpfulness of Large Language Models (LLMs), but it may lower their truthfulness. This trade-off arises because IFT steers LLMs to generate responses with long-tail knowledge that is not well covered during pre-training, leading to more informative but less truthful answers when generalizing to unseen tasks. In this paper, we empirically demonstrate this helpfulness-truthfulness trade-off in IFT and propose $\textbf{UNIT}$, a novel IFT paradigm to address it. UNIT teaches LLMs to recognize their uncertainty and explicitly reflect it at the end of their responses. Experimental results show that UNIT-tuned models maintain their helpfulness while distinguishing between certain and uncertain claims, thereby reducing hallucinations.

Large language models (LLMs) exhibit impressive performance across diverse tasks but often struggle to accurately gauge their knowledge boundaries, leading to confident yet incorrect responses. This paper explores leveraging LLMs' internal states to enhance their perception of knowledge boundaries from efficiency and risk perspectives. We investigate whether LLMs can estimate their confidence using internal states before response generation, potentially saving computational resources. Our experiments on datasets like Natural Questions, HotpotQA, and MMLU reveal that LLMs demonstrate significant pre-generation perception, which is further refined post-generation, with perception gaps remaining stable across varying conditions. To mitigate risks in critical domains, we introduce Consistency-based Confidence Calibration ($C^3$), which assesses confidence consistency through question reformulation. $C^3$ significantly improves LLMs' ability to recognize their knowledge gaps, enhancing the unknown perception rate by 5.6\% on NQ and 4.9\% on HotpotQA. Our findings suggest that pre-generation confidence estimation can optimize efficiency, while $C^3$ effectively controls output risks, advancing the reliability of LLMs in practical applications.

World models that forecast environmental changes from actions are vital for autonomous driving models with strong generalization. The prevailing driving world model mainly build on video prediction model. Although these models can produce high-fidelity video sequences with advanced diffusion-based generator, they are constrained by their predictive duration and overall generalization capabilities. In this paper, we explore to solve this problem by combining generation loss with MAE-style feature-level context learning. In particular, we instantiate this target with three key design: (1) A more scalable Diffusion Transformer (DiT) structure trained with extra mask construction task. (2) we devise diffusion-related mask tokens to deal with the fuzzy relations between mask reconstruction and generative diffusion process. (3) we extend mask construction task to spatial-temporal domain by utilizing row-wise mask for shifted self-attention rather than masked self-attention in MAE. Then, we adopt a row-wise cross-view module to align with this mask design. Based on above improvement, we propose MaskGWM: a Generalizable driving World Model embodied with Video Mask reconstruction. Our model contains two variants: MaskGWM-long, focusing on long-horizon prediction, and MaskGWM-mview, dedicated to multi-view generation. Comprehensive experiments on standard benchmarks validate the effectiveness of the proposed method, which contain normal validation of Nuscene dataset, long-horizon rollout of OpenDV-2K dataset and zero-shot validation of Waymo dataset. Quantitative metrics on these datasets show our method notably improving state-of-the-art driving world model.

Large Language Models (LLMs) have achieved remarkable success in various domains but remain vulnerable to adversarial jailbreak attacks. Existing prompt-defense strategies, including parameter-modifying and parameter-free approaches, face limitations in adaptability, interpretability, and customization, constraining their effectiveness against evolving threats. To address these challenges, we propose ShieldLearner, a novel paradigm that mimics human learning in defense. Through trial and error, it autonomously distills attack signatures into a Pattern Atlas and synthesizes defense heuristics into a Meta-analysis Framework, enabling systematic and interpretable threat detection. Furthermore, we introduce Adaptive Adversarial Augmentation to generate adversarial variations of successfully defended prompts, enabling continuous self-improvement without model retraining. In addition to standard benchmarks, we create a hard test set by curating adversarial prompts from the Wildjailbreak dataset, emphasizing more concealed malicious intent. Experimental results show that ShieldLearner achieves a significantly higher defense success rate than existing baselines on both conventional and hard test sets, while also operating with lower computational overhead, making it a practical and efficient solution for real-world adversarial defense.

Since the emergence of the Large Language Model (LLM), LLM has been widely used in fields such as writing, translating, and searching. However, there is still great potential for LLM-based methods in handling complex tasks such as decision-making in the StarCraft II environment. To address problems such as lack of relevant knowledge and poor control over subtasks of varying importance, we propose a Hierarchical Expert Prompt (HEP) for LLM. Our method improves the understanding of game situations through expert-level tactical knowledge, improving the processing quality of tasks of varying importance through a hierarchical framework. Our approach defeated the highest level (Elite) standard built-in agent in TextStarCraft II for the first time and consistently outperformed the baseline method in other difficulties. Our experiments suggest that the proposed method is a practical solution for tackling complex decision-making challenges. The replay video can be viewed on https://www.bilibili.com/video/BV1uz42187EF and https://youtu.be/dO3PshWLV5M, and our codes have been open-sourced on https://github.com/luchang1113/HEP-LLM-play-StarCraftII.

We present a method leveraging extreme learning machine (ELM) type randomized neural networks (NNs) for learning the exact time integration algorithm for initial value problems (IVPs). The exact time integration algorithm for non-autonomous systems can be represented by an algorithmic function in higher dimensions, which satisfies an associated system of partial differential equations with corresponding boundary conditions. Our method learns the algorithmic function by solving this associated system using ELM with a physics informed approach. The trained ELM network serves as the learned algorithm and can be used to solve the IVP with arbitrary initial data or step sizes from some domain. When the right hand side of the non-autonomous system exhibits a periodicity with respect to any of its arguments, while the solution itself to the problem is not periodic, we show that the algorithmic function is either periodic, or when it is not, satisfies a well-defined relation for different periods. This property can greatly simplify the algorithm learning in many problems. We consider explicit and implicit NN formulations, leading to explicit or implicit time integration algorithms, and discuss how to train the ELM network by the nonlinear least squares method. Extensive numerical experiments with benchmark problems, including non-stiff, stiff and chaotic systems, show that the learned NN algorithm produces highly accurate solutions in long-time simulations, with its time-marching errors decreasing nearly exponentially with increasing degrees of freedom in the neural network. We compare extensively the computational performance (accuracy vs.~cost) between the current NN algorithm and the leading traditional time integration algorithms. The learned NN algorithm is computationally competitive, markedly outperforming the traditional algorithms in many problems., Comment: 40 pages, 28 figures, 8 tables

Soliton microcombs generated in nonlinear microresonators facilitate the photonic integration of timing, frequency synthesis, and astronomical calibration functionalities. For these applications, low-repetition-rate soliton microcombs are essential as they establish a coherent link between optical and microwave signals. However, the required pump power typically scales with the inverse of the repetition rate, and the device footprint scales with the inverse of square of the repetition rate, rendering low-repetition-rate soliton microcombs challenging to integrate within photonic circuits. This study designs and fabricates silicon nitride microresonators on 4-inch wafers with highly compact form factors. The resonator geometries are engineered from ring to finger and spiral shapes to enhance integration density while attaining quality factors over 10^7. Driven directly by an integrated laser, soliton microcombs with repetition rates below 10 GHz are demonstrated via turnkey initiation. The phase noise performance of the synthesized microwave signals reaches -130 dBc/Hz at 100 kHz offset frequency for 10 GHz carrier frequencies. This work enables the high-density integration of soliton microcombs for chip-based microwave photonics and spectroscopy applications., Comment: 10 pages, 6 figures

This paper presents an interpretable reward design framework for reinforcement learning based constrained optimal control problems with state and terminal constraints. The problem is formalized within a standard partially observable Markov decision process framework. The reward function is constructed from four weighted components: a terminal constraint reward, a guidance reward, a penalty for state constraint violations, and a cost reduction incentive reward. A theoretically justified reward design is then presented, which establishes bounds on the weights of the components. This approach ensures that constraints are satisfied and objectives are optimized while mitigating numerical instability. Acknowledging the importance of prior knowledge in reward design, we sequentially solve two subproblems, using each solution to inform the reward design for the subsequent problem. Subsequently, we integrate reinforcement learning with curriculum learning, utilizing policies derived from simpler subproblems to assist in tackling more complex challenges, thereby facilitating convergence. The framework is evaluated against original and randomly weighted reward designs in a multi-agent particle environment. Experimental results demonstrate that the proposed approach significantly enhances satisfaction of terminal and state constraints and optimization of control cost.

Proteins are dynamic molecular machines whose biological functions, spanning enzymatic catalysis, signal transduction, and structural adaptation, are intrinsically linked to their motions. Designing proteins with targeted dynamic properties, however, remains a challenge due to the complex, degenerate relationships between sequence, structure, and molecular motion. Here, we introduce VibeGen, a generative AI framework that enables end-to-end de novo protein design conditioned on normal mode vibrations. VibeGen employs an agentic dual-model architecture, comprising a protein designer that generates sequence candidates based on specified vibrational modes and a protein predictor that evaluates their dynamic accuracy. This approach synergizes diversity, accuracy, and novelty during the design process. Via full-atom molecular simulations as direct validation, we demonstrate that the designed proteins accurately reproduce the prescribed normal mode amplitudes across the backbone while adopting various stable, functionally relevant structures. Notably, generated sequences are de novo, exhibiting no significant similarity to natural proteins, thereby expanding the accessible protein space beyond evolutionary constraints. Our work integrates protein dynamics into generative protein design, and establishes a direct, bidirectional link between sequence and vibrational behavior, unlocking new pathways for engineering biomolecules with tailored dynamical and functional properties. This framework holds broad implications for the rational design of flexible enzymes, dynamic scaffolds, and biomaterials, paving the way toward dynamics-informed AI-driven protein engineering.

The space-borne gravitational wave observatory is to detect low-frequency gravitational wave signals in the range of 0.1 mHz to 100 mHz. The inertial sensors of space gravitational wave require very high accuracy for acceleration noise, and the interaction of the space magnetic field with the test mass can generate magnetic moment forces and Lorentz forces, which lead to acceleration noise. Here, we obtain space magnetic field data from OMNI during 25 years from 1998 to 2022. And accordingly, we calculate the acceleration noise of space magnetic field of a heliocentric gravitational wave observatory, LISA, in more than 2 solar activity cycles. Then, we obtain the amplitude spectral densities of the acceleration noise for each day of the 25 years. We find that the median of the space magnetic field acceleration noise of LISA at 1 mHz is about $1 \times \rm 10^{-17}~m s^{-2}~Hz^{-1/2}$. We compare the space magnetic field acceleration noise of LISA and a geocentric gravitational wave observatory, TianQin, and find that the acceleration noise of the space magnetic field is of comparable magnitude for TianQin and LISA, and neither of them exceeds the respective acceleration noise requirements. Based on the statistical result of space magnetic field acceleration noise in more than 2 solar cycles, we give the $\chi$--$\xi$ parameters map of the TM for LISA and TianQin, and find that TianQin has a more stringent requirement of the parameters design than that of LISA.

Enhancing light-matter interactions with photonic structures is critical in classical and quantum nanophotonics. Recently, Moir\'e twisted bilayer optical materials have been proposed as a promising means towards a tunable and controllable platform for nanophotonic devices, with proof of principle realisations in the near infrared spectral range. However, the realisation of Moir\'e photonic crystal (PhC) cavities has been challenging, due to a lack of advanced nanofabrication techniques and availability of standalone transparent membranes. Here, we leverage the properties of the van der Waals material hexagonal Boron Nitride to realize Moir\'e bilayer PhC cavities. We design and fabricate a range of devices with controllable twist angles, with flatband modes in the visible spectral range (~ 450 nm). Optical characterization confirms the presence of spatially periodic cavity modes originating from the engineered dispersion relation (flatband). Our findings present a major step towards harnessing a two-dimensional van der Waals material for the next-generation of on chip, twisted nanophotonic systems.

Recent developments in machine learning (ML) techniques enable users to extract, transmit, and reproduce information semantics via ML-based semantic communication (SemCom). This significantly increases network spectral efficiency and transmission robustness. In the network, the semantic encoders and decoders among various users, based on ML, however, require collaborative updating according to new transmission tasks. The various heterogeneous characteristics of most networks in turn introduce emerging but unique challenges for semantic codec updating that are different from other general ML model updating. In this article, we first overview the key components of the SemCom system. We then discuss the unique challenges associated with semantic codec updates in heterogeneous networks. Accordingly, we point out a potential framework and discuss the pros and cons thereof. Finally, several future research directions are also discussed., Comment: 6 pages, 5 figures

Due to its rich structure and close connection with gauge theory, hyperk\"ahler manifolds have attracted increasing interest. Using infinite dimensional hyperk\"ahler reduction, Kronheimer proved that certain adjoint orbits of complexified semisimple Lie algebras admits hyperk\"ahler structures. Later on, Xu obtained a proof for the existence of hyperk\"ahler structures on adjoint orbits of $\mathfrak{sl}(2,\mathbb{C})$ from the viewpoint of symplectic geometry. This paper aims to thoroughly investigate and elucidate the key differences as well as the underlying connections between two distinct construction methods.

Consider an unmanned aerial vehicle (UAV) physically connected to the ground station with a tether operating in a space, tasked with performing precise maneuvers while constrained by the physical limitation of its tether, which prevents it from flying beyond a maximum allowable length. Violating this tether constraint could lead to system failure or operational hazards, making it essential to enforce safety constraints dynamically while ensuring the drone can track desired trajectories accurately. This paper presents a Control Barrier Function Quadratic Programming Framework (CBF-QP) for ensuring the safe and efficient operation of tethered unmanned aerial vehicles (TUAVs). The framework leverages nominal backstepping control to achieve trajectory tracking, augmented with control barrier functions to ensure compliance with the tether constraint. In this proposed method, the tether constraint is directly embedded in the control design and therefore guarantees the TUAV remains within a predefined operational region defined by the maximum tether length while achieving precise trajectory tracking. The effectiveness of the proposed framework is validated through simulations involving set-point tracking, dynamic trajectory following, and disturbances such as incorrect user inputs. The results demonstrate that the TUAV respects the tether constraint ||x(t)||

Time series analysis has witnessed the inspiring development from traditional autoregressive models, deep learning models, to recent Transformers and Large Language Models (LLMs). Efforts in leveraging vision models for time series analysis have also been made along the way but are less visible to the community due to the predominant research on sequence modeling in this domain. However, the discrepancy between continuous time series and the discrete token space of LLMs, and the challenges in explicitly modeling the correlations of variates in multivariate time series have shifted some research attentions to the equally successful Large Vision Models (LVMs) and Vision Language Models (VLMs). To fill the blank in the existing literature, this survey discusses the advantages of vision models over LLMs in time series analysis. It provides a comprehensive and in-depth overview of the existing methods, with dual views of detailed taxonomy that answer the key research questions including how to encode time series as images and how to model the imaged time series for various tasks. Additionally, we address the challenges in the pre- and post-processing steps involved in this framework and outline future directions to further advance time series analysis with vision models.

Future wireless networks must support emerging applications where environmental awareness is as critical as data transmission. Integrated Sensing and Communication (ISAC) enables this vision by allowing base stations (BSs) to allocate bandwidth and power to mobile users (MUs) for communications and cooperative sensing. However, this resource allocation is highly challenging due to: (i) dynamic resource demands from MUs and resource supply from BSs, and (ii) the selfishness of MUs and BSs. To address these challenges, existing solutions rely on either real-time (online) resource trading, which incurs high overhead and failures, or static long-term (offline) resource contracts, which lack flexibility. To overcome these limitations, we propose the Future Resource Bank for ISAC, a hybrid trading framework that integrates offline and online resource allocation through a level-wise client model, where MUs and their coalitions negotiate with BSs. We introduce two mechanisms: (i) Role-Friendly Win-Win Matching (offRFW$^2$M), leveraging overbooking to establish risk-aware, stable contracts, and (ii) Effective Backup Win-Win Matching (onEBW$^2$M), which dynamically reallocates unmet demand and surplus supply. We theoretically prove stability, individual rationality, and weak Pareto optimality of these mechanisms. Through simulations, we show that our framework improves social welfare, latency, and energy efficiency compared to existing methods.

3D vision is of paramount importance for numerous applications ranging from machine intelligence to precision metrology. Despite much recent progress, the majority of 3D imaging hardware remains bulky and complicated and provides much lower image resolution compared to their 2D counterparts. Moreover, there are many well-known scenarios that existing 3D imaging solutions frequently fail. Here, we introduce an extended monocular 3D imaging (EM3D) framework that fully exploits the vectorial wave nature of light. Via the multi-stage fusion of diffraction- and polarization-based depth cues, using a compact monocular camera equipped with a diffractive-refractive hybrid lens, we experimentally demonstrate the snapshot acquisition of a million-pixel and accurate 3D point cloud for extended scenes that are traditionally challenging, including those with low texture, being highly reflective, or nearly transparent, without a data prior. Furthermore, we discover that the combination of depth and polarization information can unlock unique new opportunities in material identification, which may further expand machine intelligence for applications like target recognition and face anti-spoofing. The straightforward yet powerful architecture thus opens up a new path for a higher-dimensional machine vision in a minimal form factor, facilitating the deployment of monocular cameras for applications in much more diverse scenarios.

We are currently in an era of escalating technological complexity and profound societal transformations, where artificial intelligence (AI) technologies exemplified by large language models (LLMs) have reignited discussions on the 'Technological Singularity'. 'Technological Singularity' is a philosophical concept referring to an irreversible and profound transformation that occurs when AI capabilities surpass those of humans comprehensively. However, quantitative modeling and analysis of the historical evolution and future trends of AI technologies remain scarce, failing to substantiate the singularity hypothesis adequately. This paper hypothesizes that the development of AI technologies could be characterized by the superposition of multiple logistic growth processes. To explore this hypothesis, we propose a multi-logistic growth process model and validate it using two real-world datasets: AI Historical Statistics and Arxiv AI Papers. Our analysis of the AI Historical Statistics dataset assesses the effectiveness of the multi-logistic model and evaluates the current and future trends in AI technology development. Additionally, cross-validation experiments on the Arxiv AI Paper, GPU Transistor and Internet User dataset enhance the robustness of our conclusions derived from the AI Historical Statistics dataset. The experimental results reveal that around 2024 marks the fastest point of the current AI wave, and the deep learning-based AI technologies are projected to decline around 2035-2040 if no fundamental technological innovation emerges. Consequently, the technological singularity appears unlikely to arrive in the foreseeable future.

Retrieval-Augmented Generation (RAG) is an advanced technique designed to address the challenges of Artificial Intelligence-Generated Content (AIGC). By integrating context retrieval into content generation, RAG provides reliable and up-to-date external knowledge, reduces hallucinations, and ensures relevant context across a wide range of tasks. However, despite RAG's success and potential, recent studies have shown that the RAG paradigm also introduces new risks, including robustness issues, privacy concerns, adversarial attacks, and accountability issues. Addressing these risks is critical for future applications of RAG systems, as they directly impact their trustworthiness. Although various methods have been developed to improve the trustworthiness of RAG methods, there is a lack of a unified perspective and framework for research in this topic. Thus, in this paper, we aim to address this gap by providing a comprehensive roadmap for developing trustworthy RAG systems. We place our discussion around five key perspectives: reliability, privacy, safety, fairness, explainability, and accountability. For each perspective, we present a general framework and taxonomy, offering a structured approach to understanding the current challenges, evaluating existing solutions, and identifying promising future research directions. To encourage broader adoption and innovation, we also highlight the downstream applications where trustworthy RAG systems have a significant impact.

Incorporating group symmetries into neural networks has been a cornerstone of success in many AI-for-science applications. Diagonal groups of isometries, which describe the invariance under a simultaneous movement of multiple objects, arise naturally in many-body quantum problems. Despite their importance, diagonal groups have received relatively little attention, as they lack a natural choice of invariant maps except in special cases. We study different ways of incorporating diagonal invariance in neural network ans\"atze trained via variational Monte Carlo methods, and consider specifically data augmentation, group averaging and canonicalization. We show that, contrary to standard ML setups, in-training symmetrization destabilizes training and can lead to worse performance. Our theoretical and numerical results indicate that this unexpected behavior may arise from a unique computational-statistical tradeoff not found in standard ML analyses of symmetrization. Meanwhile, we demonstrate that post hoc averaging is less sensitive to such tradeoffs and emerges as a simple, flexible and effective method for improving neural network solvers.

The core-collapse supernova remnant (SNR) Cassiopeia A (Cas A) is one of the brightest galactic radio sources with an angular radius of $\sim$ 2.5 $\arcmin$. Although no extension of this source has been detected in the $\gamma$-ray band, using more than 1000 days of LHAASO data above $\sim 0.8$ TeV, we find that its spectrum is significantly softer than those obtained with Imaging Air Cherenkov Telescopes (IACTs) and its flux near $\sim 1$ TeV is about two times higher. In combination with analyses of more than 16 years of \textit{Fermi}-LAT data covering $0.1 \, \mathrm{GeV} - 1 \, \mathrm{TeV}$, we find that the spectrum above 30 GeV deviates significantly from a single power-law, and is best described by a smoothly broken power-law with a spectral index of $1.90 \pm 0.15_\mathrm{stat}$ ($3.41 \pm 0.19_\mathrm{stat}$) below (above) a break energy of $0.63 \pm 0.21_\mathrm{stat} \, \mathrm{TeV}$. Given differences in the angular resolution of LHAASO-WCDA and IACTs, TeV $\gamma$-ray emission detected with LHAASO may have a significant contribution from regions surrounding the SNR illuminated by particles accelerated earlier, which, however, are treated as background by IACTs. Detailed modelling can be used to constrain acceleration processes of TeV particles in the early stage of SNR evolution.