Search

Recent development in Artificial Intelligence (AI) models has propelled their application in scientific discovery, but the validation and exploration of these discoveries require subsequent empirical experimentation. The concept of self-driving laboratories promises to automate and thus boost the experimental process following AI-driven discoveries. However, the transition of experimental protocols, originally crafted for human comprehension, into formats interpretable by machines presents significant challenges, which, within the context of specific expert domain, encompass the necessity for structured as opposed to natural language, the imperative for explicit rather than tacit knowledge, and the preservation of causality and consistency throughout protocol steps. Presently, the task of protocol translation predominantly requires the manual and labor-intensive involvement of domain experts and information technology specialists, rendering the process time-intensive. To address these issues, we propose a framework that automates the protocol translation process through a three-stage workflow, which incrementally constructs Protocol Dependence Graphs (PDGs) that approach structured on the syntax level, completed on the semantics level, and linked on the execution level. Quantitative and qualitative evaluations have demonstrated its performance at par with that of human experts, underscoring its potential to significantly expedite and democratize the process of scientific discovery by elevating the automation capabilities within self-driving laboratories., Comment: In Advances in Neural Information Processing Systems (NeurIPS'24)

In recent years, machine learning has demonstrated impressive capability in handling molecular science tasks. To support various molecular properties at scale, machine learning models are trained in the multi-task learning paradigm. Nevertheless, data of different molecular properties are often not aligned: some quantities, e.g. equilibrium structure, demand more cost to compute than others, e.g. energy, so their data are often generated by cheaper computational methods at the cost of lower accuracy, which cannot be directly overcome through multi-task learning. Moreover, it is not straightforward to leverage abundant data of other tasks to benefit a particular task. To handle such data heterogeneity challenges, we exploit the specialty of molecular tasks that there are physical laws connecting them, and design consistency training approaches that allow different tasks to exchange information directly so as to improve one another. Particularly, we demonstrate that the more accurate energy data can improve the accuracy of structure prediction. We also find that consistency training can directly leverage force and off-equilibrium structure data to improve structure prediction, demonstrating a broad capability for integrating heterogeneous data., Comment: Published as a conference paper at NeurIPS 2024

Data-driven storytelling serves as a crucial bridge for communicating ideas in a persuasive way. However, the manual creation of data stories is a multifaceted, labor-intensive, and case-specific effort, limiting their broader application. As a result, automating the creation of data stories has emerged as a significant research thrust. Despite advances in Artificial Intelligence, the systematic generation of data stories remains challenging due to their hybrid nature: they must frame a perspective based on a seed idea in a top-down manner, similar to traditional storytelling, while coherently grounding insights of given evidence in a bottom-up fashion, akin to data analysis. These dual requirements necessitate precise constraints on the permissible space of a data story. In this viewpoint, we propose integrating constraints into the data story generation process. Defined upon the hierarchies of interpretation and articulation, constraints shape both narrations and illustrations to align with seed ideas and contextualized evidence. We identify the taxonomy and required functionalities of these constraints. Although constraints can be heterogeneous and latent, we explore the potential to represent them in a computation-friendly fashion via Domain-Specific Languages. We believe that leveraging constraints will facilitate both artistic and scientific aspects of data story generation., Comment: In 2024 IEEE Visualization and Visual Analytics Gen4DS (VIS-Gen4DS'24)

Crafting automation systems tailored for specific domains requires aligning the space of human experts' semantics with the space of robot executable actions, and scheduling the required resources and system layout accordingly. Regrettably, there are three major gaps, fine-grained domain-specific knowledge injection, heterogeneity between human knowledge and robot instructions, and diversity of users' preferences, resulting automation system design a case-by-case and labour-intensive effort, thus hindering the democratization of automation. We refer to this challenging alignment as the abstract hardware grounding problem, where we firstly regard the procedural operations in humans' semantics space as the abstraction of hardware requirements, then we ground such abstractions to instantiated hardware devices, subject to constraints and preferences in the real world -- optimizing this problem is essentially standardizing and automating the design of automation systems. On this basis, we develop an automated design framework in a hybrid data-driven and principle-derived fashion. Results on designing self-driving laboratories for enhancing experiment-driven scientific discovery suggest our framework's potential to produce compact systems that fully satisfy domain-specific and user-customized requirements with no redundancy., Comment: In International Conference on Intelligent Robotics and Applications (ICIRA'24)

The accomplished functions of a variety of quantum devices are closely associated with the controlling of exciton transport. To this end we study the exciton transport of the two-dimensional system consisting of two-level multichains with various coupling configurations in a cavity. Two types of the chains are considered, including Tavis-Cummings and Su-Schrieffer-Heeger chain. Two conformations of the coupling between chains are considered, including square and triangle type. The effects of the inter-chain coupling, dimerization parameter, the cavity, the length of chains, and the number of chains on the exciton transport are in detail investigated for different coupling configurations of the multi-chain system through spectra and steady-state dynamics. The results show that in the absence of a cavity the exciton transport effciency is decided by the distribution of population of exciton on whole chains. However, when the cavity is considered the exciton transport currents and effciency of the system is controlled by the cavity-dressed energy level crossings and anticrossings near zero-energy modes, at which the coherent excitation and Landau-Zener transitions occur. Therefore, the exciton transport can be enhanced or suppressed at the crossings and anticrossings, in which the polariton acts as crucial role. Besides, it is discovered that the exciton transport effciency is closely related with the parity of both the length and the number of chains. This work is important for the understanding of the exciton transport mechanism in the multichain-cavity system, and provides theoretical basis for excitonic devices with controllable and effcient exciton transport.

We propose to effectively realize a time-dependent gravitational acceleration by using a running elevator, so that a simple pendulum inside it effectively becomes one with a time-dependent gravitational acceleration. We did such an experiment using a realistic elevator, and analyzed the data. The acceleration of an elevator is much smaller than the gravitational acceleration, and is time-dependent only when the elevator starts and stops. However, we have managed to establish the effect on the oscillation of the pendulum. The effect becomes pronounced if the simple pendulum is put in a container vertically accelerating, and the acceleration is time-dependent, while its magnitude is comparable with that of the gravitational acceleration., Comment: 6 pages

Entanglement is essential for quantum information processing but is limited by noise. We address this by developing high-yield entanglement distillation protocols with several advancements. (1) We extend the 2-to-1 recurrence entanglement distillation protocol to higher-rate n-to-(n-1) protocols that can correct any single-qubit errors. These protocols are evaluated through numerical simulations focusing on fidelity and yield. We also outline a method to adapt any classical error-correcting code for entanglement distillation, where the code can correct both bit-flip and phase-flip errors by incorporating Hadamard gates. (2) We propose a constant-depth decoder for stabilizer codes that transforms logical states into physical ones using single-qubit measurements. This decoder is applied to entanglement distillation protocols, reducing circuit depth and enabling protocols derived from advanced quantum error-correcting codes. We demonstrate this by evaluating the circuit complexity for entanglement distillation protocols based on surface codes and quantum convolutional codes. (3) Our stabilizer entanglement distillation techniques advance quantum computing. We propose a fault-tolerant protocol for constant-depth encoding and decoding of arbitrary quantum states, applicable to quantum low-density parity-check (qLDPC) codes and surface codes. This protocol is feasible with state-of-the-art reconfigurable atom arrays and surpasses the limits of conventional logarithmic depth encoders. Overall, our study integrates stabilizer formalism, measurement-based quantum computing, and entanglement distillation, advancing both quantum communication and computing., Comment: 19 pages, 7 figures

An accelerating elevator changes the apparent weight of any object inside it from the original weight, as measured inside the elevator, because the acceleration causes an inertial force on it. For any object in a running elevator, the variation of the acceleration of the elevator causes the variation of the apparent weight of the object. We have studied the time dependence of the apparent weight of the object and thus the acceleration of the elevator. For chosen initial and final floors, we measured the apparent weight of an object by using an electronic scale inside the elevator, and shot the readings of the scale and a watch during the movement of the elevator. Then we analyzed the data collected from the recorded video. If the initial and final floors are exchanged, the variations of the weight and acceleration are, respectively, same in magnitudes and opposite in signs. The experiments indicate that for the elevator to go directly from a floor to another, the process consists of periods with variable acceleration, constant acceleration, uniform motion, variable deceleration, constant deceleration and variable deceleration consecutively. If there are pauses during the movement, each pause causes an additional process consisting of periods with deceleration, stop and acceleration, replacing the original period of constant motion. Depending on the distance to the destination, the elevator reduces or diminishes the periods of constant acceleration and uniform motion., Comment: English translation of a Chinese article published in Physics Teaching (a journal of Chinese Physical Society)

High-dimensional encoding of quantum information holds the potential to greatly increase the computational power of existing devices by enlarging the accessible state space for fixed register size and by reducing the number of required entangling gates. However, qudit-based quantum computation remains far less developed than conventional qubit-based approaches, in particular for photons, which represent natural multi-level information carriers that play a crucial role in the development of quantum networks. A major obstacle for realizing quantum gates between two individual photons is the restriction of direct interaction between photons in linear media. In particular, essential logic components for quantum operations such as native qudit-qudit entangling gates are still missing for optical quantum information processing. Here we address this challenge by presenting a protocol for realizing an entangling gate -- the controlled phase-flip (CPF) gate -- for two photonic qudits in arbitrary dimension. We experimentally demonstrate this protocol by realizing a four-dimensional qudit-qudit CPF gate, whose decomposition would require at least 13 two-qubit entangling gates. Our photonic qudits are encoded in orbital angular momentum (OAM) and we have developed a new active high-precision phase-locking technology to construct a high-dimensional OAM beam splitter that increases the stability of the CPF gate, resulting in a process fidelity within a range of $ [0.64 \pm 0.01, 0.82 \pm 0.01]$. Our experiment represents a significant advance for high-dimensional optical quantum information processing and has the potential for wider applications beyond optical system., Comment: 14 pages, 7 figures

The early detection and precise diagnosis of liver tumors are tasks of critical clinical value, yet they pose significant challenges due to the high heterogeneity and variability of liver tumors. In this work, a precise LIver tumor DIAgnosis network on multi-phase contrast-enhance CT, named LIDIA, is proposed for real-world scenario. To fully utilize all available phases in contrast-enhanced CT, LIDIA first employs the iterative fusion module to aggregate variable numbers of image phases, thereby capturing the features of lesions at different phases for better tumor diagnosis. To effectively mitigate the high heterogeneity problem of liver tumors, LIDIA incorporates asymmetric contrastive learning to enhance the discriminability between different classes. To evaluate our method, we constructed a large-scale dataset comprising 1,921 patients and 8,138 lesions. LIDIA has achieved an average AUC of 93.6% across eight different types of lesions, demonstrating its effectiveness. Besides, LIDIA also demonstrated strong generalizability with an average AUC of 89.3% when tested on an external cohort of 828 patients., Comment: Accepted to MICCAI 2024

Esophageal varices (EV), a serious health concern resulting from portal hypertension, are traditionally diagnosed through invasive endoscopic procedures. Despite non-contrast computed tomography (NC-CT) imaging being a less expensive and non-invasive imaging modality, it has yet to gain full acceptance as a primary clinical diagnostic tool for EV evaluation. To overcome existing diagnostic challenges, we present the Multi-Organ-cOhesion-Network (MOON), a novel framework enhancing the analysis of critical organ features in NC-CT scans for effective assessment of EV. Drawing inspiration from the thorough assessment practices of radiologists, MOON establishes a cohesive multiorgan analysis model that unifies the imaging features of the related organs of EV, namely esophagus, liver, and spleen. This integration significantly increases the diagnostic accuracy for EV. We have compiled an extensive NC-CT dataset of 1,255 patients diagnosed with EV, spanning three grades of severity. Each case is corroborated by endoscopic diagnostic results. The efficacy of MOON has been substantiated through a validation process involving multi-fold cross-validation on 1,010 cases and an independent test on 245 cases, exhibiting superior diagnostic performance compared to methods focusing solely on the esophagus (for classifying severe grade: AUC of 0.864 versus 0.803, and for moderate to severe grades: AUC of 0.832 versus 0.793). To our knowledge, MOON is the first work to incorporate a synchronized multi-organ NC-CT analysis for EV assessment, providing a more acceptable and minimally invasive alternative for patients compared to traditional endoscopy., Comment: Early accepted to MICCAI 2024

The China Space Station Telescope (CSST, also known as Xuntian) is a serviceable two-meter-aperture wide-field telescope operating in the same orbit as the China Space Station. The CSST plans to survey a sky area of 17,500 deg$^2$ of the medium-to-high Galactic latitude to a depth of 25-26 AB mag in at least 6 photometric bands over 255-1000 nm. Within such a large sky area, slitless spectra will also be taken over the same wavelength range as the imaging survey. Even though the CSST survey is not dedicated to time-domain studies, it would still detect a large number of transients, such as supernovae (SNe). In this paper, we simulate photometric SN observations based on a strawman survey plan using the Sncosmo package. During its 10-year survey, the CSST is expected to observe about 5 million SNe of various types. With quality cuts, we obtain a "gold" sample that comprises roughly 7,400 SNe Ia, 2,200 SNe Ibc, and 6,500 SNe II candidates with correctly classified percentages reaching 91%, 63%, and 93% (formally defined as classification precision), respectively. The same survey can also trigger alerts for the detection of about 15,500 SNe Ia (precision 61%) and 2,100 SNe II (precision 49%) candidates at least two days before the light maxima. Moreover, the near-ultraviolet observations of the CSST will be able to catch hundreds of shock-cooling events serendipitously every year. These results demonstrate that the CSST can make a potentially significant contribution to SN studies., Comment: 13 pages, 9 figures, and 4 tables. Accepted for publication in SCPMA

An increasing number of experimental measurements from the BESIII, Belle, and Belle-II collaborations encourage investigations into charmed baryon two-body decay processes. By including contributions from the penguin diagrams that are ignored in previous studies, we perform a global analysis with SU(3) flavor symmetry. Assuming all form factors are real, we achieve a remarkable minimal $\chi^{2}/d.o.f = 0.788$ and find that the contribution of the amplitude proportional to $V_{cb}^*V_{ub}$ is of the order $\sim 0.01$, comparable with the contribution of the tree-level diagram. Additionally, by using the KPW theorem to reduce the number of amplitudes from 13 to 7 in the leading contribution, it becomes possible to consider the complex form factor case for the leading IRA amplitude in the global analysis. However, the analysis of complex form factors significantly conflicts with the experimental data $Br(\Xi_c^0\to\Xi^-\pi^+)$, and by excluding this data, $\chi^2/d.o.f$ is reduced from 5.95 to 1.19. Although the analysis of complex form factors shows a significant central value of the penguin diagram contribution, the large errors from the corresponding form factors make it a challenge to precisely determine its true contribution. Consequently, the direct CP violation in decay processes is predicted to be approximately zero. With more data in future experiments, the penguin diagram contribution with the amplitude proportional to $V_{cb}^*V_{ub}$ will be precisely determined, allowing for a more accurate prediction of CP violation. Our analysis necessitates further theoretical investigations and experimental measurements in the future., Comment: 13 pages, accepted by EPJC

We investigate the purely non-factorizable $\Lambda_{c}^{+}\to\Xi^{0}K^{+}$ decay using light-cone sum rules. To extract the decay amplitudes, a three-point correlation function is defined and calculated at hadron and quark-gluon level, respectively. Both the W-exchange and the W-inward emission diagrams are considered in the quark-gluon level calculation, where the two-particle light-cone distribution amplitudes (LCDAs) of kaon are used as non-perturbative input. We obtain the decay amplitudes contributed from the twist-2 and twist-3 kaon LCDAs, respectively. The obtained P-wave amplitude is consistent with those predicted in the literature, while the S-wave one is much smaller. This significant difference between the S- and P- wave amplitudes leads to small up-down spin asymmetry, closing to the experimental measurement., Comment: 13 pages, 6 figures, 3 tables

Magnetic skyrmions emerge as promising quasi-particles for encoding information in nextgeneration spintronic devices. Their innate flexibility in shape is essential for the applications although they were often ideally treated as rigid particles. In this work, we investigated the voltagecontrolled uniform strain mediated dynamics of deformed skyrmions in heterostructures with a flower-shaped magnetic nanostructure, using micromagnetic simulations. The simulated results revealed the possible states of isolated skyrmion nucleated in the nanostructure, which can be mutually switched by applying suitable in-plane strain pulses. In addition, it was found that the skyrmion motions are driven by the emerging internal forces and strain force, which originate from the asymmetric deformation of skyrmion structures. Furthermore, an analytical model of deformed skyrmions was proposed to interpret the dependences of internal forces and strain force on the asymmetric deformation of skyrmion, with some formulae derived for these forces in a semi-analytical approach. Further calculations based on these formulae verified the forces appearing in the skyrmion motion, with the resulting forces showing consistence with the simulated data. This suggested that our semi-analytical model successfully captures the main physics responsible for the motion of deformed skyrmion in the nanostructure. Our work extends the understanding of the mechanics emerging in deformed skyrmion, and provides an effective approach for deterministic manipulation of deformed skyrmion motion via strain forces and internal forces, which may be instructive to design of skyrmion-based spintronic devices.

Accurate representation of procedures in restricted scenarios, such as non-standardized scientific experiments, requires precise depiction of constraints. Unfortunately, Domain-specific Language (DSL), as an effective tool to express constraints structurally, often requires case-by-case hand-crafting, necessitating customized, labor-intensive efforts. To overcome this challenge, we introduce the AutoDSL framework to automate DSL-based constraint design across various domains. Utilizing domain specified experimental protocol corpora, AutoDSL optimizes syntactic constraints and abstracts semantic constraints. Quantitative and qualitative analyses of the DSLs designed by AutoDSL across five distinct domains highlight its potential as an auxiliary module for language models, aiming to improve procedural planning and execution., Comment: In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (ACL'24)

We present a systematic investigation of lepton pair production through photon-photon fusion processes in heavy-ion collisions. It is demonstrated that the dilepton production at a given impact parameter ($b_\perp$) with a fixed transverse momentum imbalance ($q_\perp$) can be factorized into a unified formula in terms of the Wigner photon distribution of heavy nuclei. We show that this framework provides a comprehensive description of all the relevant data from RHIC to the LHC, with a strong evidence that the quasi-real photon can be radiated not only from the nucleus as a whole, standing for the coherent contribution, but also from the sub-structures inside the nucleus, representing the incoherent contribution. Further predictions are made for the anisotropies in the correlations between $q_\perp$, $b_\perp$, and the dilepton transverse momentum ($P_\perp$). This will help us to constrain the photon Wigner distribution which plays a crucial role to study the gluonic matter of nucleus at small-$x$ through the diffractive photoproduction processes in heavy ion collision., Comment: 7 pages (main text) plus 18 pages (supplemental material), 19 figures

How to quickly solve the problem of high-order unconstrained binary optimization (HUBO) has attracted much attention, because of its importance and wide-range applications. Here we implement HUBO using a quantum adiabatic algorithm and achieve algorithmic speedup by introducing gauge symmetry into the algorithm. Gauge symmetry enforces the state to be in the instantaneous ground state, further speeding up the computation. Specifically we map the HUBO problem to quantum Z2 lattice gauge theory defined on the dual graph. The gauge operators are found by using the closed-loop-search algorithm, and subsequently the speedup scheme with gauge symmetry for HUBO problem is developed. As an example demonstrated in the classical computers, we present the mathematical formulation of our speedup scheme and propose the so-called gauged local annealing (gLQA) , which is the local quantum annealing (LQA) protected by the gauge symmetry. We then use gLQA to calculate the ground state energy of the Z2 gauge theory. gLQA reduces the computational time by one order of magnitude from that of LQA., Comment: 8 pages

Large language models (LLMs) have demonstrated impressive performance on reasoning tasks, which can be further improved through few-shot prompting techniques. However, the current evaluation primarily focuses on carefully constructed benchmarks and neglects the consideration of real-world reasoning problems that present missing and contradictory conditions, known as ill-defined problems. Our observations suggest that existing few-shot prompting techniques are ineffective in such scenarios, often providing overconfident answers or hallucination. To further study this problem, we develop a benchmark called Problems with Missing and Contradictory conditions (PMC) and introduce two novel metrics to evaluate the performance of few-shot prompting methods in these scenarios. Our analysis using the PMC benchmark reveals a trade-off dilemma between the performance of mathematical reasoning for well-defined problems and the ability to recognize ill-defined problems. To address the challenges posed by PMC, we propose a novel few-shot prompting method called SMT-LIB Prompting (SLP), which utilizes the SMT-LIB language to model the problems instead of solving them directly. Subsequently, a double-check solving strategy checks the satisfiability and uniqueness of the solution and provides final feedback. Extensive experiments demonstrate the superiority of our SLP approach compared to existing few-shot prompting methods when dealing with problems with missing and contradictory conditions. We will open-source our benchmark and code to facilitate future research., Comment: Preprint. arXiv admin note: text overlap with arXiv:2304.09797

Accurate and fast prediction of materials properties is central to the digital transformation of materials design. However, the vast design space and diverse operating conditions pose significant challenges for accurately modeling arbitrary material candidates and forecasting their properties. We present MatterSim, a deep learning model actively learned from large-scale first-principles computations, for efficient atomistic simulations at first-principles level and accurate prediction of broad material properties across the periodic table, spanning temperatures from 0 to 5000 K and pressures up to 1000 GPa. Out-of-the-box, the model serves as a machine learning force field, and shows remarkable capabilities not only in predicting ground-state material structures and energetics, but also in simulating their behavior under realistic temperatures and pressures, signifying an up to ten-fold enhancement in precision compared to the prior best-in-class. This enables MatterSim to compute materials' lattice dynamics, mechanical and thermodynamic properties, and beyond, to an accuracy comparable with first-principles methods. Specifically, MatterSim predicts Gibbs free energies for a wide range of inorganic solids with near-first-principles accuracy and achieves a 15 meV/atom resolution for temperatures up to 1000K compared with experiments. This opens an opportunity to predict experimental phase diagrams of materials at minimal computational cost. Moreover, MatterSim also serves as a platform for continuous learning and customization by integrating domain-specific data. The model can be fine-tuned for atomistic simulations at a desired level of theory or for direct structure-to-property predictions, achieving high data efficiency with a reduction in data requirements by up to 97%.

In this paper, we explore the the phase transition, critical behavior and microstructure of the FRW in the framework of a new higher order generalized uncertainty principle. Our initial step involves deriving the equation of state by defining the work density $W$ from GUP-corrected Friedmann equations as the thermodynamic pressure $P$. Based on the modified equation of state, we conduct an analysis of the $P-V$ phase transition in the FRW universe. Subsequently, we obtain the critical exponents and coexistence curves for the small and large phases of the FRW universe around the critical point. Finally, employing Ruppeiner geometry, we derive the thermodynamic curvature scalar $R_N$, investigating its sign-changing curve and spinodal curve. The results reveal distinctive thermodynamic properties for FRW universes with positive and negative GUP parameters $\beta$. In the case of $\beta>0$, the phase transition, critical behavior and microstructure of FRW universe are consistent with those of Van der Waals fluids. Conversely, for $\beta<0$, the results resemble those obtained through effective scalar field theory. These findings underscore the capacity of quantum gravity to induce phase transitions in the universe, warranting further in-depth exploration., Comment: 12 pages, 8 figures

In practice, we usually need to build variable-sized models adapting for diverse resource constraints in different application scenarios, where weight initialization is an important step prior to training. The Learngene framework, introduced recently, firstly learns one compact part termed as learngene from a large well-trained model, after which learngene is expanded to initialize variable-sized models. In this paper, we start from analysing the importance of guidance for the expansion of well-trained learngene layers, inspiring the design of a simple but highly effective Learngene approach termed SWS (Stage-wise Weight Sharing), where both learngene layers and their learning process critically contribute to providing knowledge and guidance for initializing models at varying scales. Specifically, to learn learngene layers, we build an auxiliary model comprising multiple stages where the layer weights in each stage are shared, after which we train it through distillation. Subsequently, we expand these learngene layers containing stage information at their corresponding stage to initialize models of variable depths. Extensive experiments on ImageNet-1K demonstrate that SWS achieves consistent better performance compared to many models trained from scratch, while reducing around 6.6x total training costs. In some cases, SWS performs better only after 1 epoch tuning. When initializing variable-sized models adapting for different resource constraints, SWS achieves better results while reducing around 20x parameters stored to initialize these models and around 10x pre-training costs, in contrast to the pre-training and fine-tuning approach.

Graph Neural Networks have demonstrated great success in various fields of multimedia. However, the distribution shift between the training and test data challenges the effectiveness of GNNs. To mitigate this challenge, Test-Time Training (TTT) has been proposed as a promising approach. Traditional TTT methods require a demanding unsupervised training strategy to capture the information from test to benefit the main task. Inspired by the great annotation ability of Large Language Models (LLMs) on Text-Attributed Graphs (TAGs), we propose to enhance the test-time training on graphs with LLMs as annotators. In this paper, we design a novel Test-Time Training pipeline, LLMTTT, which conducts the test-time adaptation under the annotations by LLMs on a carefully-selected node set. Specifically, LLMTTT introduces a hybrid active node selection strategy that considers not only node diversity and representativeness, but also prediction signals from the pre-trained model. Given annotations from LLMs, a two-stage training strategy is designed to tailor the test-time model with the limited and noisy labels. A theoretical analysis ensures the validity of our method and extensive experiments demonstrate that the proposed LLMTTT can achieve a significant performance improvement compared to existing Out-of-Distribution (OOD) generalization methods.

Establishing dense anatomical correspondence across distinct imaging modalities is a foundational yet challenging procedure for numerous medical image analysis studies and image-guided radiotherapy. Existing multi-modality image registration algorithms rely on statistical-based similarity measures or local structural image representations. However, the former is sensitive to locally varying noise, while the latter is not discriminative enough to cope with complex anatomical structures in multimodal scans, causing ambiguity in determining the anatomical correspondence across scans with different modalities. In this paper, we propose a modality-agnostic structural representation learning method, which leverages Deep Neighbourhood Self-similarity (DNS) and anatomy-aware contrastive learning to learn discriminative and contrast-invariance deep structural image representations (DSIR) without the need for anatomical delineations or pre-aligned training images. We evaluate our method on multiphase CT, abdomen MR-CT, and brain MR T1w-T2w registration. Comprehensive results demonstrate that our method is superior to the conventional local structural representation and statistical-based similarity measures in terms of discriminability and accuracy., Comment: Accepted by CVPR2024

Large-scale recommendation systems are characterized by their reliance on high cardinality, heterogeneous features and the need to handle tens of billions of user actions on a daily basis. Despite being trained on huge volume of data with thousands of features, most Deep Learning Recommendation Models (DLRMs) in industry fail to scale with compute. Inspired by success achieved by Transformers in language and vision domains, we revisit fundamental design choices in recommendation systems. We reformulate recommendation problems as sequential transduction tasks within a generative modeling framework ("Generative Recommenders"), and propose a new architecture, HSTU, designed for high cardinality, non-stationary streaming recommendation data. HSTU outperforms baselines over synthetic and public datasets by up to 65.8% in NDCG, and is 5.3x to 15.2x faster than FlashAttention2-based Transformers on 8192 length sequences. HSTU-based Generative Recommenders, with 1.5 trillion parameters, improve metrics in online A/B tests by 12.4% and have been deployed on multiple surfaces of a large internet platform with billions of users. More importantly, the model quality of Generative Recommenders empirically scales as a power-law of training compute across three orders of magnitude, up to GPT-3/LLaMa-2 scale, which reduces carbon footprint needed for future model developments, and further paves the way for the first foundational models in recommendations., Comment: 26 pages, 13 figures. ICML'24. Code available at https://github.com/facebookresearch/generative-recommenders

We explore the impact of initial state soft gluon radiations on the azimuthal angle asymmetries in photo-production of hard di-jet via coherent diffraction in ultraperipheral heavy ion collisions, as well as in electron-proton ($ep$) and electron-nucleus ($eA$) collisions. The primary production mechanism is identified as the diffractive production of two hard jets, accompanied by a collinear gluon emission along the beam direction. In contrast, the diffractive exclusive di-jet production, where the initial state radiation is absent, is suppressed due to color transparency. Our analysis shows that azimuthal asymmetries, traditionally attributed to final state gluon emissions, are reduced by the presence of initial state radiations. The sensitivity of azimuthal asymmetries to both initial and final state radiations suggests that they could provide novel insights into the mechanisms of di-jet production in diffractive processes., Comment: 20 pages, 5 figures

The charge density wave (CDW), as a hallmark of vanadium-based kagome superconductor AV3Sb5 (A = K, Rb, Cs), has attracted intensive attention. However, the fundamental controversy regarding the underlying mechanism of CDW therein persists. Recently, the vanadium-based bi-layered kagome metal ScV6Sn6, reported to exhibit a long-range charge order below 94 K, has emerged as a promising candidate to further clarify this core issue. Here, employing micro-focusing angle-resolved photoemission spectroscopy ({\mu}-ARPES) and first-principles calculations, we systematically studied the unique CDW order in vanadium-based bi-layered kagome metals by comparing ScV6Sn6 with its isostructural counterpart YV6Sn6, which lacks a CDW ground state. Combining ARPES data and the corresponding joint density of states (DOS), we suggest that the VHS nesting mechanism might be invalid in these materials. Besides, in ScV6Sn6, we identified multiple hybridization energy gaps resulting from CDW-induced band folding, along with an anomalous band dispersion, implying a potential electron-phonon coupling driven mechanism underlying the formation of the CDW order. Our finding not only comprehensively maps the electronic structure of V-based bi-layer kagome metals but also provide constructive experimental evidence for the unique origin of CDW in this system., Comment: 14 pages, 5 figures

We propose a theory for how the weak phonon-mediated interaction ($J_{\rm A}\!=\!1\!\sim\!4$meV) wins over the prohibitive Coulomb repulsion ($U\!=\!30\!\sim\!60$meV) and leads to a superconductor in magic-angle twisted bilayer graphene (MATBG). We find the pairing mechanism akin to that in the A$_3$C$_{60}$ family of molecular superconductors: Each AA stacking region of MATBG resembles a C$_{60}$ molecule, in that optical phonons can dynamically lift the degeneracy of the moir\'e orbitals, in analogy to the dynamical Jahn-Teller effect. Such induced $J_{\rm A}$ has the form of an inter-valley anti-Hund's coupling and is less suppressed than $U$ by the Kondo screening near a Mott insulator. Additionally, we also considered an intra-orbital Hund's coupling $J_{\rm H}$ that originates from the on-site repulsion of a carbon atom. Under a reasonable approximation of the realistic model, we prove that the renormalized local interaction between quasi-particles must have a pairing (negative) channel in a doped correlated insulator at $\nu=\pm(2+\delta\nu)$, albeit the bare interaction is positive definite. The proof is non-perturbative and based on exact asymptotic behaviors of the vertex function imposed by Ward identities. Existence of an optimal $U$ for superconductivity is predicted. We also analyzed the pairing symmetry. In a large area of the parameter space of $J_{\rm A}$, $J_{\rm H}$, the ground state has a nematic $d$-wave singlet pairing, which, however, can lead to a $p$-wave-like nodal structure due to the Berry's phase on Fermi surfaces (or Euler obstruction).

We studied the decays of Heavy baryon into a pseudoscalar meson and a dark baryon in the recently developed $B$-Mesogenesis scenario, where the two types of effective Lagrangians proposed by the scenario are both considered. The decay amplitudes of $\Lambda_b^0$ are calculated by light-cone sum rules using its light-cone distribution amplitudes. The decay amplitudes of $\Xi_b^{0,\pm}$ are related with those of $\Lambda_b^0$ through a flavor SU(3) analysis. The uncertainties of threshold parameter and the Borel parameter are both considered in the numerical calculation. The values of effective coupling constants in the $B$-Mesogenesis are taken as their upper limits that obtained from our previous study on the inclusive decay. The upper limits of the decay branching fractions are presented as functions of the dark baryon mass., Comment: 21 pages, 9 figures, 3 tables

Vector-like quarks~(VLQs) are a common feature of many scenarios of new physics beyond the Standard Model~(SM), which generally decay into a SM third-generation quark with a SM gauge boson, or a Higgs boson. The presence of a new exotic decay mode of VLQs will reduce the branching ratios of these standard decay modes and thus relax the current mass exclusion limits from LHC experiments. Based on a model-independent framework, we investigate the prospect of discovering the pair production of the weak-singlet VLQ-$B$ at the future 3-TeV Compact Linear Collider~(CLIC), by focusing on the final states including one $Z$ boson and four $b$-jets via two types of modes: $Z\to \ell^{+}\ell^{-}$ and $Z\to \nu\bar{\nu}$. By performing a rapid detector simulation of the signal and background events, and considering the initial state radiation and beamstrahlung effects, the exclusion limit at the 95\% confidence level and the $5\sigma$ discovery prospects are respectively obtained on the branching ratio of $B\to bZ$ and the VLQ-$B$ masses at the future 3-TeV CLIC with an integrated luminosity of 5 ab$^{-1}$., Comment: 17 pages, 6 figures. arXiv admin note: text overlap with arXiv:2112.15044

Motivated by recent experimental data on $\Sigma^+\to p\gamma$ at BESIII, we investigate a class of hyperon weak radiative decays. To estimate these processes, in our research, we employ a new type of light-front quark model with a three-quark picture for octet baryons. In the three-quark picture, with the use of $SU(3)_f$ and spin symmetries, we present a general form of the light front wave function for each octet baryon. By including contributions from the penguin diagram and W exchange diagram, we perform a complete calculation on the branching ratios ($Br$) and the asymmetry parameter ($\alpha$) for hyperon weak radiative decay processes. Our results are helpful for discovering additional hyperon weak radiative decay processes in experimental facilities, and our research will promote the theoretical study of baryons., Comment: 15 pages, 2 figures, 5 tables