Search

Altermagnets constitute a class of collinear compensated N\'eel ordered magnets that break time-reversal symmetry and feature spin-split band structures. Based on versatile microscopic models able to capture the altermagnetic sublattice degrees of freedom, we study characteristic local signatures of altermagnetism near disorder sites. We give a complete list of two-dimensional models that exhibit altermagnetism classified by their corresponding layer groups. Specifically, we calculate the local density of states in the vicinity of pointlike nonmagnetic impurities and expose its spatial dependence for two minimal models showcasing $d$-wave and $g$-wave altermagnetism. The momentum structure of the nodes ($d$-wave, $g$-wave, etc.) is directly imprinted on the total local density of states, thus measurable by scanning tunneling conductance experiments. This signature is present both in the spin-resolved as well as the spin-summed local density of states. We find a weaker response in the nonmagnetic state from the anisotropic crystal environment and uncover the importance of the sublattice degree of freedom to model altermagnets. We also study coexistence phases of altermagnetism and superconductivity and provide predictions for the local impurity response of in-gap bound states. The response of impurity bound states strongly enhances the distinct altermagnetic signature.

Large Language Models (LLMs) have demonstrated impressive performance across various tasks, and their application in edge scenarios has attracted significant attention. However, sparse-activated Mixture-of-Experts (MoE) models, which are well suited for edge scenarios, have received relatively little attention due to their high memory demands. Offload-based methods have been proposed to address this challenge, but they face difficulties with expert prediction. Inaccurate expert predictions can result in prolonged inference delays. To promote the application of MoE models in edge scenarios, we propose Fate, an offloading system designed for MoE models to enable efficient inference in resource-constrained environments. The key insight behind Fate is that gate inputs from adjacent layers can be effectively used for expert prefetching, achieving high prediction accuracy without additional GPU overhead. Furthermore, Fate employs a shallow-favoring expert caching strategy that increases the expert hit rate to 99\%. Additionally, Fate integrates tailored quantization strategies for cache optimization and IO efficiency. Experimental results show that, compared to Load on Demand and Expert Activation Path-based method, Fate achieves up to 4.5x and 1.9x speedups in prefill speed and up to 4.1x and 2.2x speedups in decoding speed, respectively, while maintaining inference quality. Moreover, Fate's performance improvements are scalable across different memory budgets.

Large language models (LLMs) have achieved impressive performance but face high computational costs and latency, limiting their deployment in resource-constrained settings. In contrast, small-scale LLMs (SLMs) are more efficient yet struggle to capture evolving real-world knowledge. Retrieval-augmented generation (RAG) helps by integrating external knowledge, but imperfect retrieval can introduce distracting noise that misleads SLMs. We propose RoseRAG, a robust RAG framework for SLMs via Margin-aware Preference Optimization. RoseRAG employs multi-turn prompting for detailed reasoning, rejection sampling for high-quality explanations, and contrastive preference selection to refine responses by maximizing the likelihood gap between preferred and non-preferred outputs. By integrating these components into a margin-aware optimization process, RoseRAG robustly enhances the accuracy and reliability of SLMs for RAG applications. Extensive experiments on three open-domain question answering benchmarks indicate that our innovative RoseRAG surpasses state-of-the-art baselines significantly.

Mixture of Experts (MoE), with its distinctive sparse structure, enables the scaling of language models up to trillions of parameters without significantly increasing computational costs. However, the substantial parameter size presents a challenge for inference, as the expansion in GPU memory cannot keep pace with the growth in parameters. Although offloading techniques utilise memory from the CPU and disk and parallelise the I/O and computation for efficiency, the computation for each expert in MoE models is often less than the I/O, resulting in numerous bubbles in the pipeline. Therefore, we propose Klotski, an efficient MoE inference engine that significantly reduces pipeline bubbles through a novel expert-aware multi-batch pipeline paradigm. The proposed paradigm uses batch processing to extend the computation time of the current layer to overlap with the loading time of the next layer. Although this idea has been effectively applied to dense models, more batches may activate more experts in the MoE, leading to longer loading times and more bubbles. Thus, unlike traditional approaches, we balance computation and I/O time and minimise bubbles by orchestrating their inference orders based on their heterogeneous computation and I/O requirements and activation patterns under different batch numbers. Moreover, to adapt to different hardware environments and models, we design a constraint-sensitive I/O-compute planner and a correlation-aware expert prefetcher for a schedule that minimises pipeline bubbles. Experimental results demonstrate that Klotski achieves a superior throughput-latency trade-off compared to state-of-the-art techniques, with throughput improvements of up to 85.12x.

Nonsymmorphic magnetic Weyl semimetal materials such as ReAlX (Re=rare earth, X=Si/Ge) provide a unique opportunity to explore the correlated phenomena between Weyl fermions and nontrivial magnetic configurations. To be specific, we study a lattice model in which the magnetic configuration is determined by the competition among ferromagnetic (FM) interaction, the Dzyaloshinsky-Moriya interaction, and the Kondo coupling $K_0$ to the Weyl fermion. Both quantum and finite-temperature phase transitions between FM and correlated nesting helical configurations are found. Different from the uncorrelated helimagnet that decouples to the Weyl fermions, this correlated helimagnet induces a magnetic Brillouin zone with a $K_0$-dependent nesting in the band structure of the conducting fermions instead of the magnetic monopole-like Weyl cone. By measuring the current induced by the chiral magnetic effect on the conducting fermion with nesting Weyl nodes, one can distinguish the correlated nesting helical order and the ferromagnetism because the chiral magnetic effect is considerably suppressed in the former case. These properties we study here may explain the experimental observations in ReAlX., Comment: 9 pages, 6 figures

Growing awareness of wellness has prompted people to consider whether their dietary patterns align with their health and fitness goals. In response, researchers have introduced various wearable dietary monitoring systems and dietary assessment approaches. However, these solutions are either limited to identifying foods with simple ingredients or insufficient in providing analysis of individual dietary behaviors with domain-specific knowledge. In this paper, we present DietGlance, a system that automatically monitors dietary in daily routines and delivers personalized analysis from knowledge sources. DietGlance first detects ingestive episodes from multimodal inputs using eyeglasses, capturing privacy-preserving meal images of various dishes being consumed. Based on the inferred food items and consumed quantities from these images, DietGlance further provides nutritional analysis and personalized dietary suggestions, empowered by the retrieval augmentation generation module on a reliable nutrition library. A short-term user study (N=33) and a four-week longitudinal study (N=16) demonstrate the usability and effectiveness of DietGlance.

Newton-Okounkov bodies serve as a bridge between algebraic geometry and convex geometry, enabling the application of combinatorial and geometric methods to the study of linear systems on algebraic varieties. This paper contributes to understanding the algebro-geometric information encoded in the collection of all Newton-Okounkov bodies on a given surface, focusing on Newton-Okounkov polygons with few vertices and on elliptic K3 surfaces. Let S be an algebraic surface and mv(S) be the maximum number of vertices of the Newton-Okounkov bodies of S. We prove that mv(S) = 4 if and only if its Picard number \rho(S) is at least 2 and S contains no negative irreducible curve. Additionally, if S contains a negative curve, then \rho(S) = 2 if and only if mv(S) = 5. Furthermore, we provide an example involving two elliptic K3 surfaces to demonstrate that when \rho(S) \geq 3, mv(S) no longer determines the Picard number \rho(S).

This paper introduces DeepFlow, a scalable and serverless AI platform designed to efficiently serve large language models (LLMs) at scale in cloud environments. DeepFlow addresses key challenges such as resource allocation, serving efficiency, and cold start latencies through four main design components. First, it uses a simple serverless abstraction called the request-job-task model, which helps manage AI workloads across post-training and model serving tasks. Second, it builds an in-house serving engine FlowServe using a microkernel-inspired design, NPU-centric execution, and SPMD-based parallelism to optimize LLM serving. The system also includes novel scheduling policies tailored for both PD-disaggregated and PD-colocated configurations. With optimizations like pre-warmed pods, DRAM pre-loading, and NPU-fork, DeepFlow can scale up to 64 instances in seconds. DeepFlow has been in production for over a year, operating on a large Ascend NPU cluster and providing industrystandard APIs for fine-tuning, agent serving, and model serving to our customers.

Empathetic dialogue is crucial for natural human-computer interaction, allowing the dialogue system to respond in a more personalized and emotionally aware manner, improving user satisfaction and engagement. The emergence of large language models (LLMs) has revolutionized dialogue generation by harnessing their powerful capabilities and shown its potential in multimodal domains. Many studies have integrated speech with text-based LLMs to take speech question as input and output text response. However, the lack of spoken question-answering datasets that include speech style information to supervised fine-tuning (SFT) limits the performance of these systems. As a result, while these systems excel at understanding speech content, they often struggle to generate empathetic responses. In response, we propose a novel approach that circumvents the need for question-answering data, called Listen, Perceive, and Express (LPE). Our method employs a two-stage training process, initially guiding the LLM to listen the content and perceive the emotional aspects of speech. Subsequently, we utilize Chain-of-Thought (CoT) prompting to unlock the model's potential for expressing empathetic responses based on listened spoken content and perceived emotional cues. We employ experiments to prove the effectiveness of proposed method. To our knowledge, this is the first attempt to leverage CoT for speech-based dialogue., Comment: Accepted by ICASSP 2025

We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI., Comment: A technical report from MiniMax. The authors are listed in alphabetical order. We open-sourced our MiniMax-01 at https://github.com/MiniMax-AI

Few-shot class-incremental learning (FSCIL) involves learning new classes from limited data while retaining prior knowledge, and often results in catastrophic forgetting. Existing methods either freeze backbone networks to preserve knowledge, which limits adaptability, or rely on additional modules or prompts, introducing inference overhead. To this end, we propose Continuous Knowledge-Preserving Decomposition for FSCIL (CKPD-FSCIL), a framework that decomposes a model's weights into two parts: one that compacts existing knowledge (knowledge-sensitive components) and another that carries redundant capacity to accommodate new abilities (redundant-capacity components). The decomposition is guided by a covariance matrix from replay samples, ensuring principal components align with classification abilities. During adaptation, we freeze the knowledge-sensitive components and only adapt the redundant-capacity components, fostering plasticity while minimizing interference without changing the architecture or increasing overhead. Additionally, CKPD introduces an adaptive layer selection strategy to identify layers with redundant capacity, dynamically allocating adapters. Experiments on multiple benchmarks show that CKPD-FSCIL outperforms state-of-the-art methods., Comment: Code: https://github.com/xiaojieli0903/CKPD-FSCIL

Urban segregation refers to the physical and social division of people, often driving inequalities within cities and exacerbating socioeconomic and racial tensions. While most studies focus on residential spaces, they often neglect segregation across "activity spaces" where people work, socialize, and engage in leisure. Human mobility data offers new opportunities to analyze broader segregation patterns, encompassing both residential and activity spaces, but challenges existing methods in capturing the complexity and local nuances of urban segregation. This work introduces InclusiViz, a novel visual analytics system for multi-level analysis of urban segregation, facilitating the development of targeted, data-driven interventions. Specifically, we developed a deep learning model to predict mobility patterns across social groups using environmental features, augmented with explainable AI to reveal how these features influence segregation. The system integrates innovative visualizations that allow users to explore segregation patterns from broad overviews to fine-grained detail and evaluate urban planning interventions with real-time feedback. We conducted a quantitative evaluation to validate the model's accuracy and efficiency. Two case studies and expert interviews with social scientists and urban analysts demonstrated the system's effectiveness, highlighting its potential to guide urban planning toward more inclusive cities., Comment: Accepted to IEEE TVCG (Journal Track of PacificVis 2025)

Realizing odd-parity, time-reversal-preserving, non-relativistic spin splitting is a central goal for spintronics applications. We propose a group-theory-based microscopic framework to induce odd-parity spin splitting from coplanar antiferromagnetic (AFM) states without spin-orbit coupling (SOC). We develop phenomenological models for 421 conventional period-doubling AFM systems in non-symmorphic space groups and construct minimal microscopic models for 119 of these. We find that these AFM states can attain three possible competing ground states. These ground states all break symmetries in addition to those broken by the usual AFM order. Specifically, they give rise to either odd-parity spin-splitting, nematic order, or scalar odd-parity order related to multiferroicity. Our microscopic theories reveal that the odd-parity spin-splitting energy scale is generically large and further reveal that the scalar odd-parity order gives a non-zero Berry curvature dipole without SOC. We identify 67 materials in the Magndata database for which our theory applies. We provide DFT calculations on FeSe that reveal an $h$-wave spin splitting consistent with our symmetry arguments and apply our microscopic model to determine the non-relativistic Edelstein response for CeNiAsO.

A robust $C^0$-continuous nonconforming virtual element method (VEM) is developed for a boundary value problem arising from strain gradient elasticity in two dimensions, with the family of polygonal meshes satisfying a very general geometric assumption given in Brezzi et al. (2009) and Chen and Huang (2018). The stability condition of the VEMs is derived by establishing Korn-type inequalities and inverse inequalities. Some crucial commutative relations for locking-free analysis as in elastic problems are derived. The sharp and uniform error estimates with respect to both the microscopic parameter and the Lam\'e coefficient are achieved in the lowest-order case, which is also verified by numerical results., Comment: Virtual element method for strain gradient elasticity

Recent advancements in AI alignment techniques have significantly improved the alignment of large language models (LLMs) with static human preferences. However, the dynamic nature of human preferences can render some prior training data outdated or even erroneous, ultimately causing LLMs to deviate from contemporary human preferences and societal norms. Existing methodologies, whether they involve the curation of new data for continual alignment or the manual correction of outdated data for re-alignment, demand costly human resources. To address this challenge, we propose a novel approach, Large Language Model Behavior Correction with Influence Function Recall and Post-Training (LANCET), which requires no human involvement. LANCET consists of two phases: (1) using influence functions to identify the training data that significantly impact undesirable model outputs, and (2) applying an Influence function-driven Bregman Optimization (IBO) technique to adjust the model's behavior based on these influence distributions. Our experiments demonstrate that LANCET effectively and efficiently correct inappropriate behaviors of LLMs. Furthermore, LANCET can outperform methods that rely on collecting human preferences, and it enhances the interpretability of learning human preferences within LLMs.

Low-Rank Adaptation (LoRA) is a parameter-efficient fine-tuning method that has been widely adopted in various downstream applications of LLMs. Together with the Mixture-of-Expert (MoE) technique, fine-tuning approaches have shown remarkable improvements in model capability. However, the coordination of multiple experts in existing studies solely relies on the weights assigned by the simple router function. Lack of communication and collaboration among experts exacerbate the instability of LLMs due to the imbalance load problem of MoE. To address this issue, we propose a novel MoE graph-based LLM fine-tuning framework GraphLoRA, in which a graph router function is designed to capture the collaboration signals among experts by graph neural networks (GNNs). GraphLoRA enables all experts to understand input knowledge and share information from neighbor experts by aggregating operations. Besides, to enhance each expert's capability and their collaborations, we design two novel coordination strategies: the Poisson distribution-based distinction strategy and the Normal distribution-based load balance strategy. Extensive experiments on four real-world datasets demonstrate the effectiveness of our GraphLoRA in parameter-efficient fine-tuning of LLMs, showing the benefits of facilitating collaborations of multiple experts in the graph router of GraphLoRA., Comment: 9 pages, 4 figures

As pre-trained models, like Transformers, are increasingly deployed on cloud platforms for inference services, the privacy concerns surrounding model parameters and inference data are becoming more acute. Current Privacy-Preserving Transformer Inference (PPTI) frameworks struggle with the "impossible trinity" of privacy, efficiency, and performance. For instance, Secure Multi-Party Computation (SMPC)-based solutions offer strong privacy guarantees but come with significant inference overhead and performance trade-offs. On the other hand, PPTI frameworks that use random permutations achieve inference efficiency close to that of plaintext and maintain accurate results but require exposing some model parameters and intermediate results, thereby risking substantial privacy breaches. Addressing this "impossible trinity" with a single technique proves challenging. To overcome this challenge, we propose Centaur, a novel hybrid PPTI framework. Unlike existing methods, Centaur protects model parameters with random permutations and inference data with SMPC, leveraging the structure of Transformer models. By designing a series of efficient privacy-preserving algorithms, Centaur leverages the strengths of both techniques to achieve a better balance between privacy, efficiency, and performance in PPTI. We comprehensively evaluate the effectiveness of Centaur on various types of Transformer models and datasets. Experimental results demonstrate that the privacy protection capabilities offered by Centaur can withstand various existing model inversion attack methods. In terms of performance and efficiency, Centaur not only maintains the same performance as plaintext inference but also improves inference speed by $5.0-30.4$ times.

Altermagnets break time-reversal symmetry and their spin-orbit coupling (SOC) allow for an anomalous Hall effect (AHE) that depends on the direction of the N\'eel ordering vector. The AHE and the ferromagnetic spin moment share the same symmetry and hence are usually proportional. However, density functional theory (DFT) calculations find that the AHE exists with negligible ferromagnetic spin moment for some compounds, whereas it reaches sizable values for other altermagnets. By examining realistic minimal models for altermagnetism in which the DFT phenomenology is captured, we uncover a general SOC-enabled quasi-symmetry that provides a natural explanation for the amplitude of the ferromagnetic spin moment across the vast range of different altermagnetic materials. Additionally, we derive analytic expressions for the magnetic anisotropy energy, providing a simple means to identify the preferred N\'eel vector orientation for altermagnets., Comment: 5 pages, 2 figures

Ensembles of generative large language models (LLMs) can integrate the strengths of different LLMs to compensate for the limitations of individual models. However, recent work has focused on training an additional fusion model to combine complete responses from multiple LLMs, failing to tap into their collaborative potential to generate higher-quality responses. Moreover, as the additional fusion model is trained on a specialized dataset, these methods struggle with generalizing to open-domain queries from online users. In this paper, we propose SpecFuse, a novel ensemble framework that outputs the fused result by iteratively producing the next segment through collaboration among LLMs. This is achieved through cyclic execution of its inference and verification components. In each round, the inference component invokes each base LLM to generate candidate segments in parallel, and the verify component calls these LLMs again to predict the ranking of the segments. The top-ranked segment is then broadcast to all LLMs, encouraging them to generate higher-quality segments in the next round. This approach also allows the base LLMs to be plug-and-play, without any training or adaptation, avoiding generalization limitations. Furthermore, to conserve computational resources, we propose a model exit mechanism that dynamically excludes models exhibiting poor performance in previous rounds during each query response. In this way, it effectively reduces the number of model calls while maintaining overall performance., Comment: 15 pages, 5 figures

Long-lived mechanical oscillators are actively pursued as critical resources for quantum storage, sensing, and transduction. However, achieving deterministic quantum control while limiting mechanical dissipation remains a persistent challenge. Here, we demonstrate strong coupling between a transmon superconducting qubit and an ultra-long-lived nanomechanical oscillator ($T_\text{1} \approx 25 \text{ ms}$ at 5 GHz, $Q \approx 0.8 \times 10^9$) by leveraging the low acoustic loss in silicon and phononic bandgap engineering. The qubit-oscillator system achieves large cooperativity ($C_{T_1}\approx 1.5\times10^5$, $C_{T_2}\approx 150$), enabling the generation of non-classical states and the investigation of mechanisms underlying mechanical decoherence. We show that dynamical decoupling$\unicode{x2014}$implemented through the qubit$\unicode{x2014}$can mitigate decoherence, leading to a mechanical coherence time of $T_2\approx 1 \text{ ms}$. These findings extend the exceptional storage capabilities of mechanical oscillators to the quantum regime, putting them forward as compact bosonic elements for future applications in quantum computing and metrology., Comment: 30 pages, 21 figures

Understanding the intricate magnetic structures of materials is crucial for the advancements in material science, spintronics, and geology. Recent developments of quantum-enabled magnetometers, such as nitrogen-vacancy (NV) centers in diamond, have enabled direct imaging of magnetic field distributions associated with a wide variety of magnetic profiles. However, reconstructing the magnetization of a sample from a measured magnetic field map poses a complex inverse problem, which can be further complicated by the presence of measurement noise, finite spatial resolution, and the sample-to-sensor distance. While magnetic field distributions are typically measured through various available imaging methods, accurately reconstructing the underlying structures from these maps remains a challenge. In this work, we introduce a novel and efficient GPU-based method for reconstructing magnetic source quantities (i.e. spatially varying magnetization density) from measured magnetic fields with minimum prior assumptions. We validate our method by simulating diverse magnetic structures under realistic experimental conditions, including multi-domain ferromagnetism and magnetic spin textures such as skyrmion, anti-skyrmion, and meron. Experimentally, we apply our technique to investigate the magnetic field maps from a lunar rock (Apollo lunar mare basalt sample 10003,184) obtained using a wide-field quantum diamond microscope. The reconstructed magnetic domains of the lunar rock are consistent with current paleomagnetic knowledge. Our approach provides a versatile and universal tool for investigating complex magnetization profiles, paving the way for future quantum sensing experiments., Comment: 8 pages, 4 figures

Reward modeling is crucial for aligning large language models (LLMs) with human preferences, especially in reinforcement learning from human feedback (RLHF). However, current reward models mainly produce scalar scores and struggle to incorporate critiques in a natural language format. We hypothesize that predicting both critiques and the scalar reward would improve reward modeling ability. Motivated by this, we propose Critic-RM, a framework that improves reward models using self-generated critiques without extra supervision. Critic-RM employs a two-stage process: generating and filtering high-quality critiques, followed by joint fine-tuning on reward prediction and critique generation. Experiments across benchmarks show that Critic-RM improves reward modeling accuracy by 3.7%-7.3% compared to standard reward models and LLM judges, demonstrating strong performance and data efficiency. Additional studies further validate the effectiveness of generated critiques in rectifying flawed reasoning steps with 2.5%-3.2% gains in improving reasoning accuracy., Comment: Accepted to NAACL 2025 (Main Conference)

Darboux transformation plays a key role in constructing explicit closed-form solutions of completely integrable systems. This paper provides an algebraic construction of generalized Darboux matrices with the same poles for the $2\times2$ Lax pair, in which the coefficient matrices are polynomials of spectral parameter. The first-order monic Darboux matrix is constructed explicitly and its classification theorem is presented. Then by using the solutions of the corresponding adjoint Lax pair, the $n$-order monic Darboux matrix and its inverse, both sharing the same unique pole, are derived explicitly. Further, a theorem is proposed to describe the invariance of Darboux matrix regarding pole distributions in Darboux matrix and its inverse. Finally, a unified theorem is offered to construct formal Darboux transformation in general form. All Darboux matrices expressible as the product of $n$ first-order monic Darboux matrices can be constructed in this way. The nonlocal focusing NLS equation, the focusing NLS equation and the Kaup-Boussinesq equation are taken as examples to illustrate the application of these Darboux transformations.

Code snippet adaptation is a fundamental activity in the software development process. Unlike code generation, code snippet adaptation is not a "free creation", which requires developers to tailor a given code snippet in order to fit specific requirements and the code context. Recently, large language models (LLMs) have confirmed their effectiveness in the code generation task with promising results. However, their performance on adaptation, a reuse-oriented and context-dependent code change prediction task, is still unclear. To bridge this gap, we conduct an empirical study to investigate the performance and issues of LLMs on the adaptation task. We first evaluate the adaptation performances of three popular LLMs and compare them to the code generation task. Our result indicates that their adaptation ability is weaker than generation, with a nearly 15% decrease on pass@1 and more context-related errors. By manually inspecting 200 cases, we further investigate the causes of LLMs' sub-optimal performance, which can be classified into three categories, i.e., Unclear Requirement, Requirement Misalignment and Context Misapplication. Based on the above empirical research, we propose an interactive prompting approach to eliciting LLMs' adaptation ability. Experimental result reveals that our approach greatly improve LLMs' adaptation performance. The best-performing Human-LLM interaction successfully solves 159 out of the 202 identified defects and improves the pass@1 and pass@5 by over 40% compared to the initial instruction-based prompt. Considering human efforts, we suggest multi-agent interaction as a trade-off, which can achieve comparable performance with excellent generalization ability. We deem that our approach could provide methodological assistance for autonomous code snippet reuse and adaptation with LLMs., Comment: 12 pages, 10 figures, accepted by ICSE 2025

Image editing involves a variety of complex tasks and requires efficient and precise manipulation techniques. In this paper, we present MagicQuill, an integrated image editing system that enables swift actualization of creative ideas. Our system features a streamlined yet functionally robust interface, allowing for the articulation of editing operations (e.g., inserting elements, erasing objects, altering color) with minimal input. These interactions are monitored by a multimodal large language model (MLLM) to anticipate editing intentions in real time, bypassing the need for explicit prompt entry. Finally, we apply a powerful diffusion prior, enhanced by a carefully learned two-branch plug-in module, to process editing requests with precise control. Experimental results demonstrate the effectiveness of MagicQuill in achieving high-quality image edits. Please visit https://magic-quill.github.io to try out our system., Comment: Code and demo available at https://magic-quill.github.io

Photon transport through a diffusing slab can be described by the radiative transfer equation (RTE). When the slab is highly scattering and weakly absorbing, the RTE simplifies to the diffusion equation. In this paper, an inverse diffusion approximation (IDA) method is numerically developed to determine the optical properties (reduced scattering $\mu'_s$ and absorption coefficient $\mu_a$) of a homogeneous slab using simple reflectance and transmission measurements with a spectrometer. The reflectance and transmission of a diffusing slab with an arbitrary thickness can then be predicted by solving the forward problem by using the calculated $\mu'_s$ and $\mu_a$. Our method is validated both numerically, by directly comparing with Monte-Carlo simulations, and experimentally, by comparing with measurements on ZnO/PDMS and TiO$_2$/PDMS composite polymer films with varying thicknesses. The IDA method is also applied to distinguish between different types of tissue. Overall, our method could be used to guide the design of radiative cooling reflectors, or lighting and optical display diffusers for applications in medical imaging and other fields.

Serverless computing has gained significant traction for machine learning inference applications, which are often deployed as serverless workflows consisting of multiple CPU and GPU functions with data dependency. However, existing data-passing solutions for serverless computing primarily reply on host memory for fast data transfer, mandating substantial data movement and resulting in salient I/O overhead. In this paper, we present FaaSTube, a GPU-efficient data passing system for serverless inference. FaaSTube manages intermediate data within a GPU memory pool to facilitate direct data exchange between GPU functions. It enables fine-grained bandwidth sharing over PCIe and NVLink, minimizing data-passing latency for both host-to-GPU and GPU-to-GPU while providing performance isolation between functions. Additionally, FaaSTube implements an elastic GPU memory pool that dynamically scales to accommodate varying data-passing demands. Evaluations on real-world applications show that FaaSTube reduces end-to-end latency by up to 90\% and achieves up to 12x higher throughput compared to the state-of-the-art.

Large Language Models (LLMs), such as ChatGPT, Phi3 and Llama-3, are leading a significant leap in AI, as they can generalize knowledge from their training to new tasks without fine-tuning. However, their application in the financial domain remains relatively limited. The financial field is inherently complex, requiring a deep understanding across various perspectives, from macro, micro economic trend to quantitative analysis. Motivated by this complexity, a mixture of expert LLMs tailored to specific financial domains could offer a more comprehensive understanding for intricate financial tasks. In this paper, we present the FinTeamExperts, a role-specialized LLM framework structured as a Mixture of Experts (MOEs) for financial analysis. The framework simulates a collaborative team setting by training each model to specialize in distinct roles: Macro Analysts, Micro analysts, and Quantitative Analysts. This role-specific specialization enhances the model's ability to integrate their domain-specific expertise. We achieve this by training three 8-billion parameter models on different corpus, each dedicated to excelling in specific finance-related roles. We then instruct-tune FinTeamExperts on downstream tasks to align with practical financial tasks. The experimental results show that FinTeamExperts outperform all models of the same size and larger on three out of four datasets. On the fourth dataset, which presents a more complex task, FinTeamExperts still surpass all models of the same size. This highlights the success of our role-based specialization approach and the continued training approach for FinTeamExperts.

Nonlocal, integral operators have become an efficient surrogate for bottom-up homogenization, due to their ability to represent long-range dependence and multiscale effects. However, the nonlocal homogenized model has unavoidable discrepancy from the microscale model. Such errors accumulate and propagate in long-term simulations, making the resultant prediction unreliable. To develop a robust and reliable bottom-up homogenization framework, we propose a new framework, which we coin Embedded Nonlocal Operator Regression (ENOR), to learn a nonlocal homogenized surrogate model and its structural model error. This framework provides discrepancy-adaptive uncertainty quantification for homogenized material response predictions in long-term simulations. The method is built on Nonlocal Operator Regression (NOR), an optimization-based nonlocal kernel learning approach, together with an embedded model error term in the trainable kernel. Then, Bayesian inference is employed to infer the model error term parameters together with the kernel parameters. To make the problem computationally feasible, we use a multilevel delayed acceptance Markov chain Monte Carlo (MLDA-MCMC) method, enabling efficient Bayesian model calibration and model error estimation. We apply this technique to predict long-term wave propagation in a heterogeneous one-dimensional bar, and compare its performance with additive noise models. Owing to its ability to capture model error, the learned ENOR achieves improved estimation of posterior predictive uncertainty.

Modern-world robotics involves complex environments where multiple autonomous agents must interact with each other and other humans. This necessitates advanced interactive multi-agent motion planning techniques. Generalized Nash equilibrium(GNE), a solution concept in constrained game theory, provides a mathematical model to predict the outcome of interactive motion planning, where each agent needs to account for other agents in the environment. However, in practice, multiple local GNEs may exist. Finding a single GNE itself is complex as it requires solving coupled constrained optimal control problems. Furthermore, finding all such local GNEs requires exploring the solution space of GNEs, which is a challenging task. This work proposes the MultiNash-PF framework to efficiently compute multiple local GNEs in constrained trajectory games. Potential games are a class of games for which a local GNE of a trajectory game can be found by solving a single constrained optimal control problem. We propose MultiNash-PF that integrates the potential game approach with implicit particle filtering, a sample-efficient method for non-convex trajectory optimization. We first formulate the underlying game as a constrained potential game and then utilize the implicit particle filtering to identify the coarse estimates of multiple local minimizers of the game's potential function. MultiNash-PF then refines these estimates with optimization solvers, obtaining different local GNEs. We show through numerical simulations that MultiNash-PF reduces computation time by up to 50\% compared to a baseline approach.

Neural operators (NOs) have demonstrated remarkable success in learning mappings between function spaces, serving as efficient approximators for the forward solutions of complex physical systems governed by partial differential equations (PDEs). However, while effective as black-box solvers, they offer limited insight into the underlying physical mechanism, due to the lack of interpretable representations of the physical parameters that drive the system. To tackle this challenge, we propose a new paradigm for learning disentangled representations from neural operator parameters, thereby effectively solving an inverse problem. Specifically, we introduce DisentangO, a novel hyper-neural operator architecture designed to unveil and disentangle the latent physical factors of variation embedded within the black-box neural operator parameters. At the core of DisentangO is a multi-task neural operator architecture that distills the varying parameters of the governing PDE through a task-wise adaptive layer, coupled with a hierarchical variational autoencoder that disentangles these variations into identifiable latent factors. By learning these disentangled representations, our model not only enhances physical interpretability but also enables more robust generalization across diverse physical systems. Empirical evaluations across supervised, semi-supervised, and unsupervised learning contexts show that DisentangO effectively extracts meaningful and interpretable latent features, bridging the divide between predictive performance and physical understanding in neural operator frameworks.

In this paper, we initiate the study of \emph{multi-designated detector watermarking (MDDW)} for large language models (LLMs). This technique allows model providers to generate watermarked outputs from LLMs with two key properties: (i) only specific, possibly multiple, designated detectors can identify the watermarks, and (ii) there is no perceptible degradation in the output quality for ordinary users. We formalize the security definitions for MDDW and present a framework for constructing MDDW for any LLM using multi-designated verifier signatures (MDVS). Recognizing the significant economic value of LLM outputs, we introduce claimability as an optional security feature for MDDW, enabling model providers to assert ownership of LLM outputs within designated-detector settings. To support claimable MDDW, we propose a generic transformation converting any MDVS to a claimable MDVS. Our implementation of the MDDW scheme highlights its advanced functionalities and flexibility over existing methods, with satisfactory performance metrics.

Nonconvex trajectory optimization is at the core of designing trajectories for complex autonomous systems. A challenge for nonconvex trajectory optimization methods, such as sequential convex programming, is to find an effective warm-starting point to approximate the nonconvex optimization with a sequence of convex ones. We introduce a first-order method with filter-based warm-starting for nonconvex trajectory optimization. The idea is to first generate sampled trajectories using constraint-aware particle filtering, which solves the problem as an estimation problem. We then identify different locally optimal trajectories through agglomerative hierarchical clustering. Finally, we choose the best locally optimal trajectory to warm-start the prox-linear method, a first-order method with guaranteed convergence. We demonstrate the proposed method on a multi-agent trajectory optimization problem with linear dynamics and nonconvex collision avoidance. Compared with sequential quadratic programming and interior-point method, the proposed method reduces the objective function value by up to approximately 96\% within the same amount of time for a two-agent problem, and 98\% for a six-agent problem.

Optical beam steering enables optical detection and imaging in macroscopic or microscopic scales and long-range communication over free space. It underpins numerous optical applications, including LiDAR, biomedical imaging, and remote sensing. Despite the inherent speed of light, advanced applications increasingly require the ability to steer multiple beams simultaneously to increase imaging throughput, boost communication bandwidth, and control arrays qubits for scalable quantum computing. Therefore, there is a significant demand for non-mechanical, integrated, and scalable multi-beam steering technology. Here, we report a scalable multi-beam steering system comprising an array of acousto-optic beam steering channels and photonic integrated circuits on a thin-film lithium niobate platform. Each channel generates tens of individually controllable beams of visible wavelength by exciting acoustic waves using digitally synthesized multi-tone microwave signals. We demonstrate the system's capabilities through multi-input, multi-output free-space communications, simultaneously transmitting to multiple receivers at megabits/sec data rates. This technology can be readily scaled up to steer hundreds of optical beams from a compact chip, potentially advancing many areas of optical technologies and enabling novel applications., Comment: 17 pages, 4 figures

The stability of the integrated photonic circuits is of critical importance for many applications that require high frequency precision or robust operation over time, such as optomechanical sensing, frequency conversion, optical communication, and quantum optics. Photonic memory is useful for low-energy optical computing and interconnects. Thin film lithium niobate (TFLN), as an emerging photonic platform, exhibits complex material properties including pyroelectric (PE) and photorefractive (PR) effects which could lead to intra-device drift and excess noise under different environmental or operating conditions as well as be utilized for building photonic memory. However, the long-term stability and memory effect of its optical properties has not been explored. In this paper, we discovered a long-lived change of optical refractive index as a result of light excitation and temporal temperature variation using Z-cut TFLN microresonators and reveal a strong dependence of the instability with the crystal orientation of the thin film form. The recovery time are measured to be over 10 hours. Leveraging the photonic memory with a long relaxation time, we realize optical trimming of the cavity resonance frequencies. Our result offers insights towards understanding the fundamental noise properties and dynamic behavior of the integrated TFLN material and devices., Comment: 9 pages, 5 figures

Horticulture Luminescent Solar Concentrators (HLSCs) represent an innovative concept developed in recent years to promote crop yields, building upon the foundation of traditional Luminescent Solar Concentrators (LSCs). HLSCs are characterized by two distinct properties: spectral conversion and light extraction. Unlike traditional LSCs, HLSCs focus on converting energy from one part of the solar spectrum (typically green) to a specific range (usually red) and aim for the converted photons to exit the device from the bottom surface rather than the edge surfaces. In this study, we start by examining the specific requirements of horticulture to clarify the motivation for using HLSCs. We re-evaluate and propose new optical metrics tailored to HLSCs. Additionally, we analyse potential loss channels for direct red emission and converted red emission. Utilizing Monte Carlo ray tracing method and experimental data, we further explore the factors influencing these loss channels. Our work provides a fundamental discussion on HLSCs and offers design guidelines for future HLSC research.

With the expansion of business scenarios, real recommender systems are facing challenges in dealing with the constantly emerging new tasks in multi-task learning frameworks. In this paper, we attempt to improve the generalization ability of multi-task recommendations when dealing with new tasks. We find that joint training will enhance the performance of the new task but always negatively impact existing tasks in most multi-task learning methods. Besides, such a re-training mechanism with new tasks increases the training costs, limiting the generalization ability of multi-task recommendation models. Based on this consideration, we aim to design a suitable sharing mechanism among different tasks while maintaining joint optimization efficiency in new task learning. A novel two-stage prompt-tuning MTL framework (MPT-Rec) is proposed to address task irrelevance and training efficiency problems in multi-task recommender systems. Specifically, we disentangle the task-specific and task-sharing information in the multi-task pre-training stage, then use task-aware prompts to transfer knowledge from other tasks to the new task effectively. By freezing parameters in the pre-training tasks, MPT-Rec solves the negative impacts that may be brought by the new task and greatly reduces the training costs. Extensive experiments on three real-world datasets show the effectiveness of our proposed multi-task learning framework. MPT-Rec achieves the best performance compared to the SOTA multi-task learning method. Besides, it maintains comparable model performance but vastly improves the training efficiency (i.e., with up to 10% parameters in the full training way) in the new task learning.

Recent advances in spin transport research have highlighted the potential of quantum paramagnets as platforms for exploring novel phenomena and developing next-generation technologies. In this paper, we investigate the flavor Nernst effect (FNE) in quantum paramagnets, focusing on the Hall-type thermal spin transport of crystal electric field (CEF) excitations with spin-orbit couplings. As a proof of principle, we investigate the quantum paramagnetic ground state in an effective spin-1 Hamiltonian with Dzyaloshinskii-Moriya interactions and a large hard-axis anisotropy. We employ linear flavor-wave theory to analyze the low-energy excitations, and obtain the flavor Nernst coefficients from the linear response theory. We demonstrate the FNE in a 2D pyrochlore thin film with an all-in-all-out Ising axis configuration, and investigate their dependence on temperature, anisotropy, DM interaction, and external fields. Our results reveal the connection between the FNE and the Berry curvature of the CEF excitations, suggesting potential applications in manipulating thermal spin currents and exploring topological spin transport phenomena in quantum paramagnets., Comment: 12 pages, 10 figures

Graph neural networks (GNNs) are vulnerable to adversarial attacks, especially for topology perturbations, and many methods that improve the robustness of GNNs have received considerable attention. Recently, we have witnessed the significant success of large language models (LLMs), leading many to explore the great potential of LLMs on GNNs. However, they mainly focus on improving the performance of GNNs by utilizing LLMs to enhance the node features. Therefore, we ask: Will the robustness of GNNs also be enhanced with the powerful understanding and inference capabilities of LLMs? By presenting the empirical results, we find that despite that LLMs can improve the robustness of GNNs, there is still an average decrease of 23.1% in accuracy, implying that the GNNs remain extremely vulnerable against topology attacks. Therefore, another question is how to extend the capabilities of LLMs on graph adversarial robustness. In this paper, we propose an LLM-based robust graph structure inference framework, LLM4RGNN, which distills the inference capabilities of GPT-4 into a local LLM for identifying malicious edges and an LM-based edge predictor for finding missing important edges, so as to recover a robust graph structure. Extensive experiments demonstrate that LLM4RGNN consistently improves the robustness across various GNNs. Even in some cases where the perturbation ratio increases to 40%, the accuracy of GNNs is still better than that on the clean graph. The source code can be found in https://github.com/zhongjian-zhang/LLM4RGNN., Comment: accepted by KDD 2025

Despite the recent popularity of attention-based neural architectures in core AI fields like natural language processing (NLP) and computer vision (CV), their potential in modeling complex physical systems remains under-explored. Learning problems in physical systems are often characterized as discovering operators that map between function spaces based on a few instances of function pairs. This task frequently presents a severely ill-posed PDE inverse problem. In this work, we propose a novel neural operator architecture based on the attention mechanism, which we coin Nonlocal Attention Operator (NAO), and explore its capability towards developing a foundation physical model. In particular, we show that the attention mechanism is equivalent to a double integral operator that enables nonlocal interactions among spatial tokens, with a data-dependent kernel characterizing the inverse mapping from data to the hidden parameter field of the underlying operator. As such, the attention mechanism extracts global prior information from training data generated by multiple systems, and suggests the exploratory space in the form of a nonlinear kernel map. Consequently, NAO can address ill-posedness and rank deficiency in inverse PDE problems by encoding regularization and achieving generalizability. We empirically demonstrate the advantages of NAO over baseline neural models in terms of generalizability to unseen data resolutions and system states. Our work not only suggests a novel neural operator architecture for learning interpretable foundation models of physical systems, but also offers a new perspective towards understanding the attention mechanism.

In a Fermi system near or at half-filling, a specific superconducting pairing channel, if not explicitly included in the Hamiltonian, can be boosted by suppressing a competing pairing channel; this is exemplified by the enhancement of extended $s$-wave correlations upon suppressing $s$-wave Cooper pairing. This phenomenon, originally found by the use of generalized uncertainty relations is referred to as \emph{order by projection}. The case of zero on-site Coulomb interaction in the thermodynamic limit, confirms this mechanism through the analytical solution. In this study, we go further and systematically investigate this mechanism for a strongly correlated fermionic Hubbard model, now with finite on-site interaction, on a square lattice with an extended set of hopping parameters. We explore the behaviors of different pairing channels when one of them is suppressed, utilizing density matrix renormalization group calculations. Our findings provide numerical evidence supporting the existence of \emph{order by projection} in the strongly correlated system we studied. We also investigate the effect of the strength of Hubbard $U$, next-nearest neighbor $t'$, hole-doping, as well as finite-size scaling approaching the thermodynamic limit.

Ref.[BCOW17] introduced a pioneering quantum approach (coined BCOW algorithm) for solving linear differential equations with optimal error tolerance. Originally designed for a specific class of diagonalizable linear differential equations, the algorithm was extended by Krovi in [Kro23] to encompass broader classes, including non-diagonalizable and even singular matrices. Despite the common misconception, the original algorithm is indeed applicable to non-diagonalizable matrices, with diagonalisation primarily serving for theoretical analyses to establish bounds on condition number and solution error. By leveraging basic estimates from [Kro23], we derive bounds comparable to those outlined in the Krovi algorithm, thereby reinstating the advantages of the BCOW approach. Furthermore, we extend the BCOW algorithm to address time-dependent linear differential equations by transforming non-autonomous systems into higher-dimensional autonomous ones, a technique also applicable for the Krovi algorithm., Comment: quantum algorithm for ODEs

This paper explores the explicit design of quantum circuits for quantum simulation of partial differential equations (PDEs) with physical boundary conditions. These equations and/or their discretized forms usually do not evolve via unitary dynamics, thus are not suitable for quantum simulation. Boundary conditions (either time-dependent or independent) make the problem more difficult. To tackle this challenge, the Schrodingerisation method can be employed, which converts linear partial and ordinary differential equations with non-unitary dynamics into systems of Schrodinger-type equations, via the so-called warped phase transformation that maps the equation into one higher dimension. Despite advancements in Schrodingerisation techniques, the explicit implementation of quantum circuits for solving general PDEs, especially with physical boundary conditions, remains underdeveloped. We present two methods for handling the inhomogeneous terms arising from time-dependent physical boundary conditions. One approach utilizes Duhamel's principle to express the solution in integral form and employs linear combination of unitaries (LCU) for coherent state preparation. Another method applies an augmentation to transform the inhomogeneous problem into a homogeneous one. We then apply the quantum simulation technique from [CJL23] to transform the resulting non-autonomous system to an autonomous system in one higher dimension. We provide detailed implementations of these two methods and conduct a comprehensive complexity analysis in terms of queries to the time evolution input oracle., Comment: Introductory content may overlap with our previous works, arXiv:2404.13585 and arXiv:2305.02710

Topological quantum computation based on Majorana objects is subject to a significant challenge because at least some of the two-qubit quantum gates rely on the fermion (either charge or spin) parity of the qubits. This dependency renders the quantum operations involving these gates probabilistic when attempting to advance quantum processes within the quantum circuit model. Such an approach leads to significant information loss whenever measurements yield the undesired fermion parity. To resolve the problem of wasting information, we devise topological operations that allow for the non-dissipative correction of information from undesired fermion parity to the desired one. We will use the sparse-dense mixed encoding process for the controlled-NOT gate as an example to explain how corrections can be implemented without affecting the quantum information carried by the computational qubits. This correction process can be applied {to} either the undesired input qubits or the fermion parity-dependent quantum gates, and it works for both Majorana-zero-mode-based and Majorana-edge-mode-based topological quantum computation., Comment: 9 pages, 5 figures, accepted by PRA

Large language models (LLMs) bear promise as a fast and accurate material modeling paradigm for evaluation, analysis, and design. Their vast number of trainable parameters necessitates a wealth of data to achieve accuracy and mitigate overfitting. However, experimental measurements are often limited and costly to obtain in sufficient quantities for finetuning. To this end, we present a physics-based training pipeline that tackles the pathology of data scarcity. The core enabler is a physics-based modeling framework that generates a multitude of synthetic data to align the LLM to a physically consistent initial state before finetuning. Our framework features a two-phase training strategy: (1) utilizing the large-in-amount while less accurate synthetic data for supervised pretraining, and (2) finetuning the phase-1 model with limited experimental data. We empirically demonstrate that supervised pretraining is vital to obtaining accurate finetuned LLMs, via the lens of learning polymer flammability metrics where cone calorimeter data is sparse.

Large language models (LLMs) typically utilize the top-k contexts from a retriever in retrieval-augmented generation (RAG). In this work, we propose a novel instruction fine-tuning framework RankRAG, which instruction-tunes a single LLM for the dual purpose of context ranking and answer generation in RAG. In particular, the instruction-tuned LLMs work surprisingly well by adding a small fraction of ranking data into the training blend, and outperform existing expert ranking models, including the same LLM exclusively fine-tuned on a large amount of ranking data. For generation, we compare our model with many strong baselines, including GPT-4-0613, GPT-4-turbo-2024-0409, and ChatQA-1.5, an open-sourced model with the state-of-the-art performance on RAG benchmarks. Specifically, our Llama3-RankRAG significantly outperforms Llama3-ChatQA-1.5 and GPT-4 models on nine knowledge-intensive benchmarks. In addition, it also performs comparably to GPT-4 on five RAG benchmarks in the biomedical domain without instruction fine-tuning on biomedical data, demonstrating its superb capability for generalization to new domains.

Molecular discovery, when formulated as an optimization problem, presents significant computational challenges because optimization objectives can be non-differentiable. Evolutionary Algorithms (EAs), often used to optimize black-box objectives in molecular discovery, traverse chemical space by performing random mutations and crossovers, leading to a large number of expensive objective evaluations. In this work, we ameliorate this shortcoming by incorporating chemistry-aware Large Language Models (LLMs) into EAs. Namely, we redesign crossover and mutation operations in EAs using LLMs trained on large corpora of chemical information. We perform extensive empirical studies on both commercial and open-source models on multiple tasks involving property optimization, molecular rediscovery, and structure-based drug design, demonstrating that the joint usage of LLMs with EAs yields superior performance over all baseline models across single- and multi-objective settings. We demonstrate that our algorithm improves both the quality of the final solution and convergence speed, thereby reducing the number of required objective evaluations. Our code is available at http://github.com/zoom-wang112358/MOLLEO

Recently, large code generation models trained in a self-supervised manner on extensive unlabeled programming language data have achieved remarkable success. While these models acquire vast amounts of code knowledge, they perform poorly on code understanding tasks, such as code search and clone detection, as they are specifically trained for generation. Pre-training a larger encoder-only architecture model from scratch on massive code data can improve understanding performance. However, this approach is costly and time-consuming, making it suboptimal. In this paper, we pioneer the transfer of knowledge from pre-trained code generation models to code understanding tasks, significantly reducing training costs. We examine effective strategies for enabling decoder-only models to acquire robust code representations. Furthermore, we introduce CL4D, a contrastive learning method designed to enhance the representation capabilities of decoder-only models. Comprehensive experiments demonstrate that our approach achieves state-of-the-art performance in understanding tasks such as code search and clone detection. Our analysis shows that our method effectively reduces the distance between semantically identical samples in the representation space. These findings suggest the potential for unifying code understanding and generation tasks using a decoder-only structured model., Comment: 8 pages, 4 figures