Search

The realization of fault-tolerant quantum computers hinges on effective quantum error correction protocols, whose performance significantly relies on the nature of the underlying noise. In this work, we directly study the structure of non-Markovian correlated errors and their impact on surface code memory performance. Specifically, we compare surface code performance under non-Markovian noise and independent circuit-level noise, while keeping marginal error rates constant. Our analysis shows that while not all temporally correlated structures are detrimental, certain structures, particularly multi-time "streaky" correlations affecting syndrome qubits and two-qubit gates, can severely degrade logical error rate scaling. Furthermore, we discuss our results in the context of recent quantum error correction experiments on physical devices. These findings underscore the importance of understanding and mitigating non-Markovian noise toward achieving practical, fault-tolerant quantum computing., Comment: 17 pages, 9 figures

It is a classical result that the set $K\backslash G /B$ is finite, where $G$ is a reductive algebraic group over an algebraically closed field with characteristic not equal to two, $B$ is a Borel subgroup of $G$, and $K = G^{\theta}$ is the fixed point subgroup of an involution of $G$. In this paper, we investigate the affine counterpart of the aforementioned set, where $G$ is the general linear group over formal Laurent series, $B$ is an Iwahori subgroup of $G$, and $K$ is either the orthogonal group or the symplectic group over formal Laurent series. We construct explicit bijections between the double cosets $K \backslash G/B$ and certain twisted affine involutions. This is the first combinatorial description of $K$-orbits in the affine flag variety of type A., Comment: 30 pages, 1 table

Supervised fine-tuning (SFT) is a common method to enhance the tool calling capabilities of Large Language Models (LLMs), with the training data often being synthesized. The current data synthesis process generally involves sampling a set of tools, formulating a requirement based on these tools, and generating the call statements. However, tools sampled randomly lack relevance, making them difficult to combine and thus reducing the diversity of the data. Additionally, current work overlooks the coherence between turns of dialogues, leading to a gap between the synthesized data and real-world scenarios. To address these issues, we propose a Graph-based Sampling strategy to sample more relevant tool combinations, and a Planned-generation strategy to create plans that guide the synthesis of coherent dialogues. We integrate these two strategies and enable multiple agents to synthesize the dialogue data interactively, resulting in our tool-calling data synthesis pipeline ToolFlow. Data quality assessments demonstrate improvements in the naturalness and coherence of our synthesized dialogues. Finally, we apply SFT on LLaMA-3.1-8B using 8,000 synthetic dialogues generated with ToolFlow. Results show that the model achieves tool-calling performance comparable to or even surpassing GPT-4, while maintaining strong general capabilities.

Large language models (LLMs) can store a significant amount of factual knowledge in their parameters. However, their parametric knowledge may conflict with the information provided in the context. Such conflicts can lead to undesirable model behaviour, such as reliance on outdated or incorrect information. In this work, we investigate whether LLMs can identify knowledge conflicts and whether it is possible to know which source of knowledge the model will rely on by analysing the residual stream of the LLM. Through probing tasks, we find that LLMs can internally register the signal of knowledge conflict in the residual stream, which can be accurately detected by probing the intermediate model activations. This allows us to detect conflicts within the residual stream before generating the answers without modifying the input or model parameters. Moreover, we find that the residual stream shows significantly different patterns when the model relies on contextual knowledge versus parametric knowledge to resolve conflicts. This pattern can be employed to estimate the behaviour of LLMs when conflict happens and prevent unexpected answers before producing the answers. Our analysis offers insights into how LLMs internally manage knowledge conflicts and provides a foundation for developing methods to control the knowledge selection processes., Comment: Foundation Model Interventions Workshop @ NeurIPS 2024

Large language models (LLMs) can store a significant amount of factual knowledge in their parameters. However, their parametric knowledge may conflict with the information provided in the context -- this phenomenon, known as \emph{context-memory knowledge conflicts}, can lead to undesirable model behaviour, such as reliance on outdated or incorrect information. Analysing the internal activations of LLMs, we find that they can internally register the signals of knowledge conflict at mid-layers. Such signals allow us to detect whether a knowledge conflict occurs and use \emph{inference-time} intervention strategies to resolve it. In this work, we propose \textsc{SpARE}, a \emph{training-free} representation engineering method that uses pre-trained sparse auto-encoders (SAEs) to control the knowledge selection behaviour of LLMs. \textsc{SpARE} identifies the functional features that control the knowledge selection behaviours and applies them to edit the internal activations of LLMs at inference time. Our experimental results show that \textsc{SpARE} can effectively control the usage of either knowledge source to resolve knowledge conflict in open-domain question-answering tasks, surpassing existing representation engineering methods ($+10\%$) as well as contrastive decoding methods ($+15\%$).

The tendency of Large Language Models (LLMs) to generate hallucinations raises concerns regarding their reliability. Therefore, confidence estimations indicating the extent of trustworthiness of the generations become essential. However, current LLM confidence estimations in languages other than English remain underexplored. This paper addresses this gap by introducing a comprehensive investigation of Multilingual Confidence estimation (MlingConf) on LLMs, focusing on both language-agnostic (LA) and language-specific (LS) tasks to explore the performance and language dominance effects of multilingual confidence estimations on different tasks. The benchmark comprises four meticulously checked and human-evaluate high-quality multilingual datasets for LA tasks and one for the LS task tailored to specific social, cultural, and geographical contexts of a language. Our experiments reveal that on LA tasks English exhibits notable linguistic dominance in confidence estimations than other languages, while on LS tasks, using question-related language to prompt LLMs demonstrates better linguistic dominance in multilingual confidence estimations. The phenomena inspire a simple yet effective native-tone prompting strategy by employing language-specific prompts for LS tasks, effectively improving LLMs' reliability and accuracy on LS tasks., Comment: Comments: This work was intended as a replacement of arXiv:2402.13606 and any subsequent updates will appear there

In Federated Learning (FL), anomaly detection (AD) is a challenging task due to the decentralized nature of data and the presence of non-IID data distributions. This study introduces a novel federated threshold calculation method that leverages summary statistics from both normal and anomalous data to improve the accuracy and robustness of anomaly detection using autoencoders (AE) in a federated setting. Our approach aggregates local summary statistics across clients to compute a global threshold that optimally separates anomalies from normal data while ensuring privacy preservation. We conducted extensive experiments using publicly available datasets, including Credit Card Fraud Detection, Shuttle, and Covertype, under various data distribution scenarios. The results demonstrate that our method consistently outperforms existing federated and local threshold calculation techniques, particularly in handling non-IID data distributions. This study also explores the impact of different data distribution scenarios and the number of clients on the performance of federated anomaly detection. Our findings highlight the potential of using summary statistics for threshold calculation in improving the scalability and accuracy of federated anomaly detection systems.

Large Language Models (LLMs) can interact with the real world by connecting with versatile external APIs, resulting in better problem-solving and task automation capabilities. Previous research primarily focuses on APIs with limited arguments from a single source or overlooks the complex dependency relationship between different APIs. However, it is essential to utilize multiple APIs collaboratively from various sources (e.g., different Apps in the iPhone), especially for complex user instructions. In this paper, we introduce \texttt{AppBench}, the first benchmark to evaluate LLMs' ability to plan and execute multiple APIs from various sources in order to complete the user's task. Specifically, we consider two significant challenges in multiple APIs: \textit{1) graph structures:} some APIs can be executed independently while others need to be executed one by one, resulting in graph-like execution order; and \textit{2) permission constraints:} which source is authorized to execute the API call. We have experimental results on 9 distinct LLMs; e.g., GPT-4o achieves only a 2.0\% success rate at the most complex instruction, revealing that the existing state-of-the-art LLMs still cannot perform well in this situation even with the help of in-context learning and finetuning. Our code and data are publicly available at https://github.com/ruleGreen/AppBench.

Vascular anastomosis, the surgical connection of blood vessels, is essential in procedures such as organ transplants and reconstructive surgeries. The precision required limits accessibility due to the extensive training needed, with manual suturing leading to variable outcomes and revision rates up to 7.9%. Existing robotic systems, while promising, are either fully teleoperated or lack the capabilities necessary for autonomous vascular anastomosis. We present the Micro Smart Tissue Autonomous Robot (micro-STAR), an autonomous robotic system designed to perform vascular anastomosis on small-diameter vessels. The micro-STAR system integrates a novel suturing tool equipped with Optical Coherence Tomography (OCT) fiber-optic sensor and a microcamera, enabling real-time tissue detection and classification. Our system autonomously places sutures and manipulates tissue with minimal human intervention. In an ex vivo study, micro-STAR achieved outcomes competitive with experienced surgeons in terms of leak pressure, lumen reduction, and suture placement variation, completing 90% of sutures without human intervention. This represents the first instance of a robotic system autonomously performing vascular anastomosis on real tissue, offering significant potential for improving surgical precision and expanding access to high-quality care., Comment: This paper was submitted to IEEE TMRB and is currently under review. There are 9 pages, 9 figures, and 2 tables

Since the first detection of gravitational waves (GW) in 2015, there have been significant efforts to enhance the sensitivity of the ground-based detectors within the LIGO-Virgo-KAGRA collaboration. Despite these efforts, many gravitational wave signals remain below the detection threshold due to non-linear noise from the physical environment of the detectors, coupled with the main GW signal. \texttt{DeepClean}'s capability to handle a variety of noise sources, including linear, non-linear, and \emph{non-stationary} coupling mechanisms, is particularly important in fields like astrophysics and gravitational wave astronomy, where accurate data analysis is crucial for the detection and study of gravitational waves from astronomical events., Comment: This paper needs an internal review by LIGO-Virgo collaboration before proceeding

In this study, we examined the impact of recommendation systems' algorithms on individuals' collaborator choices when forming teams. Different algorithmic designs can lead individuals to select one collaborator over another, thereby shaping their teams' composition, dynamics, and performance. To test this hypothesis, we conducted a 2 x 2 between-subject laboratory experiment with 332 participants who assembled teams using a recommendation system. We tested four algorithms that controlled the participants' agency to choose collaborators and the inclusion of fairness criteria. Our results show that participants assigned by an algorithm to work in highly diverse teams struggled to work with different and unfamiliar individuals, while participants enabled by an algorithm to choose collaborators without fairness criteria formed homogenous teams without the necessary skills. In contrast, combining users' agency and fairness criteria in an algorithm enhanced teams' performance and composition. This study breaks new ground by providing insights into how algorithms can augment team formation.

Computing-in-memory (CIM) is an emerging computing paradigm, offering noteworthy potential for accelerating neural networks with high parallelism, low latency, and energy efficiency compared to conventional von Neumann architectures. However, existing research has primarily focused on hardware architecture and network co-design for large-scale neural networks, without considering resource constraints. In this study, we aim to develop edge-friendly deep neural networks (DNNs) for accelerators based on resistive random-access memory (RRAM). To achieve this, we propose an edge compilation and resource-constrained RRAM-aware neural architecture search (NAS) framework to search for optimized neural networks meeting specific hardware constraints. Our compilation approach integrates layer partitioning, duplication, and network packing to maximize the utilization of computation units. The resulting network architecture can be optimized for either high accuracy or low latency using a one-shot neural network approach with Pareto optimality achieved through the Non-dominated Sorted Genetic Algorithm II (NSGA-II). The compilation of mobile-friendly networks, like Squeezenet and MobilenetV3 small can achieve over 80% of utilization and over 6x speedup compared to ISAAC-like framework with different crossbar resources. The resulting model from NAS optimized for speed achieved 5x-30x speedup. The code for this paper is available at https://github.com/ArChiiii/rram_nas_comp_pack., Comment: The 42nd IEEE International Conference on Computer Design (ICCD 2024)

Progress in Text-to-Image (T2I) models has significantly improved the generation of images from textual descriptions. However, existing evaluation metrics do not adequately assess the models' ability to handle a diverse range of textual prompts, which is crucial for their generalizability. To address this, we introduce a new metric called Visual Language Evaluation Understudy (VLEU). VLEU uses large language models to sample from the visual text domain, the set of all possible input texts for T2I models, to generate a wide variety of prompts. The images generated from these prompts are evaluated based on their alignment with the input text using the CLIP model.VLEU quantifies a model's generalizability by computing the Kullback-Leibler divergence between the marginal distribution of the visual text and the conditional distribution of the images generated by the model. This metric provides a quantitative way to compare different T2I models and track improvements during model finetuning. Our experiments demonstrate the effectiveness of VLEU in evaluating the generalization capability of various T2I models, positioning it as an essential metric for future research in text-to-image synthesis., Comment: accepted by EMNLP2024(long paper,main conference)

Macroscopic properties of equity markets affect the performance of active equity strategies but many are not adequately captured by conventional models of financial mathematics and econometrics. Using the CRSP Database of the US equity market, we study empirically several macroscopic properties defined in terms of market capitalizations and returns, and highlight a list of stylized facts and open questions motivated in part by stochastic portfolio theory. Additionally, we present a systematic backtest of the diversity-weighted portfolio under various configurations and study its performance in relation to macroscopic quantities. All of our results can be replicated using codes made available on our GitHub repository., Comment: 33 Pages, 15 Figures

Rural America has long been perceived as "left behind" by policies that leave it struggling while benefiting bustling urban cities and suburbs. That narrative holds some truth: rural America has a declining population, along with lower educational attainment and lower workforce participation than urban America. But while rural America certainly faces challenges, it also has its own strengths and assets. "Small Towns, Big Opportunities: Many Workers in Rural Areas Have Good Jobs, but These Areas Need Greater Investment in Education, Training, and Career Counseling" counters some of the negative stereotypes and finds that working adults in rural America are almost as likely (50 percent) as working adults in urban America (54 percent) to have a good job.

A common trend in simulation-driven engineering applications is the ever-increasing size and complexity of the problem, where classical numerical methods typically suffer from significant computational time and huge memory cost. Methods based on artificial intelligence have been extensively investigated to accelerate partial differential equations (PDE) solvers using data-driven surrogates. However, most data-driven surrogates require an extremely large amount of training data. In this paper, we propose the Convolutional Hierarchical Deep Learning Neural Network-Tensor Decomposition (C-HiDeNN-TD) method, which can directly obtain surrogate models by solving large-scale space-time PDE without generating any offline training data. We compare the performance of the proposed method against classical numerical methods for extremely large-scale systems.

Accurate and reliable prediction has profound implications to a wide range of applications. In this study, we focus on an instance of spatio-temporal learning problem--traffic prediction--to demonstrate an advanced deep learning model developed for making accurate and reliable forecast. Despite the significant progress in traffic prediction, limited studies have incorporated both explicit and implicit traffic patterns simultaneously to improve prediction performance. Meanwhile, the variability nature of traffic states necessitates quantifying the uncertainty of model predictions in a statistically principled way; however, extant studies offer no provable guarantee on the statistical validity of confidence intervals in reflecting its actual likelihood of containing the ground truth. In this paper, we propose an end-to-end traffic prediction framework that leverages three primary components to generate accurate and reliable traffic predictions: dynamic causal structure learning for discovering implicit traffic patterns from massive traffic data, causally-aware spatio-temporal multi-graph convolution network (CASTMGCN) for learning spatio-temporal dependencies, and conformal prediction for uncertainty quantification. CASTMGCN fuses several graphs that characterize different important aspects of traffic networks and an auxiliary graph that captures the effect of exogenous factors on the road network. On this basis, a conformal prediction approach tailored to spatio-temporal data is further developed for quantifying the uncertainty in node-wise traffic predictions over varying prediction horizons. Experimental results on two real-world traffic datasets demonstrate that the proposed method outperforms several state-of-the-art models in prediction accuracy; moreover, it generates more efficient prediction regions than other methods while strictly satisfying the statistical validity in coverage.

This paper presents a tutorial of an explainable approach using Convolutional Neural Network (CNN) and Gradient-weighted Class Activation Mapping (Grad-CAM) to classify four progressive dementia stages based on open MRI brain images. The detailed implementation steps are demonstrated with an explanation. Whilst the proposed CNN architecture is demonstrated to achieve more than 99% accuracy for the test dataset, the computational procedure of CNN remains a black box. The visualisation based on Grad-CAM is attempted to explain such very high accuracy and may provide useful information for physicians. Future motivation based on this work is discussed., Comment: 15 pages, 11 figures, 3 tables

We report the first observation of the nearby giant radio galaxy NGC 315 using a global VLBI array consisting of 22 radio antennas located across five continents, including high-sensitivity stations, at 22 GHz. Utilizing the extensive $(u,v)$-coverage provided by the array, coupled with the application of a recently developed super-resolution imaging technique based on the regularized maximum likelihood method, we were able to transversely resolve the NGC 315 jet at parsec scales for the first time. Previously known for its central ridge-brightened morphology at similar scales in former VLBI studies, the jet now clearly exhibits a limb-brightened structure. This finding suggests an inherent limb-brightening that was not observable before due to limited angular resolution. Considering that the jet is viewed at an angle of $\sim50^\circ$, the observed limb-brightening is challenging to reconcile with the magnetohydrodynamic models and simulations, which predict that the Doppler-boosted jet edges should dominate over the non-boosted central layer. The conventional jet model that proposes a fast spine and a slow sheath with uniform transverse emissivity may pertain to our observations. However, in this model, the relativistic spine would need to travel at speeds of $\Gamma\gtrsim6.0-12.9$ along the de-projected jet distance of (2.3-10.8) $\times 10^3$ gravitational radii from the black hole. We propose an alternative scenario that suggests higher emissivity at the jet boundary layer, resulting from more efficient particle acceleration or mass loading onto the jet edges, and consider prospects for future observations with even higher angular resolution., Comment: 25 pages, 12 figures, 1 table, accepted for publication in the Astrophysical Journal Letters

Laboratory automation is an expensive and complicated endeavor with limited inflexible options for small-scale labs. We develop a prototype system for tending to a bench-top centrifuge using computer vision methods for color detection and circular Hough Transforms to detect and localize centrifuge buckets. Initial results show that the prototype is capable of automating the usage of regular bench-top lab equipment.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

This paper introduces a new theoretical framework for optimizing second-order behaviors of wireless networks. Unlike existing techniques for network utility maximization, which only consider first-order statistics, this framework models every random process by its mean and temporal variance. The inclusion of temporal variance makes this framework well-suited for modeling Markovian fading wireless channels and emerging network performance metrics such as age-of-information (AoI) and timely-throughput. Using this framework, we sharply characterize the second-order capacity region of wireless access networks. We also propose a simple scheduling policy and prove that it can achieve every interior point in the second-order capacity region. To demonstrate the utility of this framework, we apply it to an unsolved network optimization problem where some clients wish to minimize AoI while others wish to maximize timely-throughput. We show that this framework accurately characterizes AoI and timely-throughput. Moreover, it leads to a tractable scheduling policy that outperforms other existing work., Comment: To appear in IEEE/ACM Transactions on Networking. arXiv admin note: substantial text overlap with arXiv:2201.06486

This paper propels creative control in generative visual AI by allowing users to "select". Departing from traditional text or sketch-based methods, we for the first time allow users to choose visual concepts by parts for their creative endeavors. The outcome is fine-grained generation that precisely captures selected visual concepts, ensuring a holistically faithful and plausible result. To achieve this, we first parse objects into parts through unsupervised feature clustering. Then, we encode parts into text tokens and introduce an entropy-based normalized attention loss that operates on them. This loss design enables our model to learn generic prior topology knowledge about object's part composition, and further generalize to novel part compositions to ensure the generation looks holistically faithful. Lastly, we employ a bottleneck encoder to project the part tokens. This not only enhances fidelity but also accelerates learning, by leveraging shared knowledge and facilitating information exchange among instances. Visual results in the paper and supplementary material showcase the compelling power of PartCraft in crafting highly customized, innovative creations, exemplified by the "charming" and creative birds. Code is released at https://github.com/kamwoh/partcraft., Comment: ECCV 2024. arXiv admin note: substantial text overlap with arXiv:2311.15477

Quantum teleportation is not merely a fascinating corollary of quantum entanglement, it also finds utility in quantum processing and circuit compilation. In this paper, we measure and track the entanglement and fidelity of two-qubit states prepared on a 127-qubit IBM Quantum device, as one of the qubits is teleported across 19 qubits. We design, evaluate and compare two distinct approaches to teleportation: post-selected measurement categorisation and dynamic circuit corrections based on mid-circuit measurements, and compare with direct state transportation using SWAP gates. By optimally choosing the teleportation path which exhibits the highest total negativity entanglement measure across nearest-neighbour pairs, we show the entanglement of a two-qubit graph state is sustained after at least 19 hops in teleportation using the post-selection approach and 17 hops using the dynamic circuit approach. We observe a higher level of teleported entanglement in paths determined from two-qubit negativities compared to those obtained from gate errors, demonstrating an advantage in using the negativity map over the gate error map for compiling quantum circuits.

In causal inference, estimating Heterogeneous Treatment Effects (HTEs) from observational data is critical for understanding how different subgroups respond to treatments, with broad applications such as precision medicine and targeted advertising. However, existing work on HTE, subgroup discovery, and causal visualization is insufficient to address two challenges: first, the sheer number of potential subgroups and the necessity to balance multiple objectives (e.g., high effects and low variances) pose a considerable analytical challenge. Second, effective subgroup analysis has to follow the analysis goal specified by users and provide causal results with verification. To this end, we propose a visual analytics approach for subgroup-based causal heterogeneity exploration. Specifically, we first formulate causal subgroup discovery as a constrained multi-objective optimization problem and adopt a heuristic genetic algorithm to learn the Pareto front of optimal subgroups described by interpretable rules. Combining with this model, we develop a prototype system, CausalPrism, that incorporates tabular visualization, multi-attribute rankings, and uncertainty plots to support users in interactively exploring and sorting subgroups and explaining treatment effects. Quantitative experiments validate that the proposed model can efficiently mine causal subgroups that outperform state-of-the-art HTE and subgroup discovery methods, and case studies and expert interviews demonstrate the effectiveness and usability of the system. Code is available at https://osf.io/jaqmf/?view_only=ac9575209945476b955bf829c85196e9., Comment: 12 pages, 7 figures

The terahertz (THz) frequency range, bridging the gap between microwave and infrared frequencies, presents unparalleled opportunities for advanced imaging, sensing, communications, and spectroscopy applications. Terahertz photonics, in analogy with microwave photonics, is a promising solution to address the critical challenges in THz technologies through optical methods. Despite its vast potential, key technical challenges remain in effectively interfacing THz signals with the optical domain, especially THz-optic modulation and optical generation of THz waves. Here, we address these challenges using a monolithic integrated photonic chip designed to support efficient bidirectional interaction between THz and optical waves. Leveraging the significant second-order optical nonlinearity and strong optical and THz confinement in a thin-film lithium niobate on quartz platform, the chip supports both efficient THz-optic modulation and continuous THz wave generation at up to 500 GHz. The THz-optic modulator features a radio frequency (RF) half-wave voltage of 8V at 500 GHz, representing more than an order of magnitude reduction in modulation power consumption from previous works. The measured continuous wave THz generation efficiency of 4.8*10-6 /W at 500 GHz also marks a tenfold improvement over existing tunable THz generation devices based on lithium niobate. We further leverage the coherent nature of the optical THz generation process and mature optical modulation techniques to realize high-speed electro-THz modulation at frequencies up to 35 GHz. The chip-scale THz-photonic platform paves the way for more compact, efficient, and cost-effective THz systems with potential applications in THz communications, remote sensing, and spectroscopy.

Current research found the issue of Early Answering in large language models (LLMs), where the models already have an answer before generating the Chain-of-Thought (CoT). This phenomenon suggests a potential lack of necessary dependency between the predicted answer and the reasoning process. Consequently, two important questions arise: (1) Is CoT still necessary if the model already has an answer? (2) Can the correctness of the answer serve as valid evidence for the correctness of CoT? To address these questions, we propose a method, namely Chain-of-Probe (CoP), to probe changes in the mind during the model's reasoning. The probing results show that in a significant number of question-answer cases, CoT appears to be unnecessary, and this necessity correlates with the simplicity of the task, defined by reasoning steps required. Furthermore, by analyzing patterns in mind change, we examine the correctness of the model's reasoning. Our validation reveals that many responses, although correct in their final answer, contain errors in their reasoning process. To this end, we propose a strategic approach based on CoP to prioritize answers with correct reasoning among multiple candidates, thereby bolstering the reliability of the model's reasoning.

We discuss anyon condensation in mixed-state topological order. The phases were recently conjectured to be classified by pre-modular fusion categories. Just like anyon condensation in pure-state topological order, a bootstrap analysis shows condensable anyons are given by connected \'etale algebras. We explain how to perform generic anyon condensation including non-invertible anyons and successive condensations. Interestingly, some condensations lead to pure-state topological orders. We clarify when this happens. We also compute topological invariants of equivalence classes., Comment: 52 pages, 14 figures

Large Language Models (LLMs) have shown great potential in the biomedical domain with the advancement of retrieval-augmented generation (RAG). However, existing retrieval-augmented approaches face challenges in addressing diverse queries and documents, particularly for medical knowledge queries, resulting in sub-optimal performance. To address these limitations, we propose a novel plug-and-play LLM-based retrieval method called Self-Rewarding Tree Search (SeRTS) based on Monte Carlo Tree Search (MCTS) and a self-rewarding paradigm. By combining the reasoning capabilities of LLMs with the effectiveness of tree search, SeRTS boosts the zero-shot performance of retrieving high-quality and informative results for RAG. We further enhance retrieval performance by fine-tuning LLMs with Proximal Policy Optimization (PPO) objectives using the trajectories collected by SeRTS as feedback. Controlled experiments using the BioASQ-QA dataset with GPT-3.5-Turbo and LLama2-7b demonstrate that our method significantly improves the performance of the BM25 retriever and surpasses the strong baseline of self-reflection in both efficiency and scalability. Moreover, SeRTS generates higher-quality feedback for PPO training than self-reflection. Our proposed method effectively adapts LLMs to document retrieval tasks, enhancing their ability to retrieve highly relevant documents for RAG in the context of medical knowledge queries. This work presents a significant step forward in leveraging LLMs for accurate and comprehensive biomedical question answering., Comment: This work has been accepted by EMNLP 2024

In this paper we present the European Low Frequency Survey (ELFS), a project that will enable foregrounds-free measurements of the primordial $B$-mode polarization and a detection of the tensor-to-scalar ratio, $r$, to a level $\sigma(r) = 0.001$ by measuring the Galactic and extra-galactic emissions in the 5--120\,GHz frequency window. Indeed, the main difficulty in measuring the B-mode polarization comes from the fact that many other processes in the Universe also emit polarized microwaves, which obscure the faint Cosmic Microwave Background (CMB) signal. The first stage of this project is being carried out in synergy with the Simons Array (SA) collaboration, installing a 5.5--11\,GHz (X-band) coherent receiver at the focus of one of the three 3.5\,m SA telescopes in Atacama, Chile, followed by the installation of the QUIJOTE-MFI2 in the 10--20 GHz range. We designate this initial iteration of the ELFS program as ELFS-SA. The receivers are equipped with a fully digital back-end that will provide a frequency resolution of 1\,MHz across the band, allowing us to clean the scientific signal from unwanted radio frequency interference, particularly from low-Earth orbit satellite mega constellations. This paper reviews the scientific motivation for ELFS and its instrumental characteristics, and provides an update on the development of ELFS-SA., Comment: SPIE conference on Astronomical Telescopes + Instrumentation, Yokohama, 16-22 June 2024. New version with correction in Eq. (3) arXiv admin note: text overlap with arXiv:2310.16509

Memes, which rapidly disseminate personal opinions and positions across the internet, also pose significant challenges in propagating social bias and prejudice. This study presents a novel approach to detecting harmful memes, particularly within the multicultural and multilingual context of Singapore. Our methodology integrates image captioning, Optical Character Recognition (OCR), and Large Language Model (LLM) analysis to comprehensively understand and classify harmful memes. Utilizing the BLIP model for image captioning, PP-OCR and TrOCR for text recognition across multiple languages, and the Qwen LLM for nuanced language understanding, our system is capable of identifying harmful content in memes created in English, Chinese, Malay, and Tamil. To enhance the system's performance, we fine-tuned our approach by leveraging additional data labeled using GPT-4V, aiming to distill the understanding capability of GPT-4V for harmful memes to our system. Our framework achieves top-1 at the public leaderboard of the Online Safety Prize Challenge hosted by AI Singapore, with the AUROC as 0.7749 and accuracy as 0.7087, significantly ahead of the other teams. Notably, our approach outperforms previous benchmarks, with FLAVA achieving an AUROC of 0.5695 and VisualBERT an AUROC of 0.5561.

We prove the Sato--Tate distribution of Kloosterman sums over function fields with explicit error terms, when the places vary in arithmetic progressions or short intervals. A joint Sato--Tate distribution of two ``different" exponential sums is also proved.

We prove an Erd\H{o}s--Tur\'an type inequality for compact Lie groups, from which we deduce an effective version of Deligne's equidistribution theorem., Comment: Improve the estimates in the previous version

Existing fine-grained hashing methods typically lack code interpretability as they compute hash code bits holistically using both global and local features. To address this limitation, we propose ConceptHash, a novel method that achieves sub-code level interpretability. In ConceptHash, each sub-code corresponds to a human-understandable concept, such as an object part, and these concepts are automatically discovered without human annotations. Specifically, we leverage a Vision Transformer architecture and introduce concept tokens as visual prompts, along with image patch tokens as model inputs. Each concept is then mapped to a specific sub-code at the model output, providing natural sub-code interpretability. To capture subtle visual differences among highly similar sub-categories (e.g., bird species), we incorporate language guidance to ensure that the learned hash codes are distinguishable within fine-grained object classes while maintaining semantic alignment. This approach allows us to develop hash codes that exhibit similarity within families of species while remaining distinct from species in other families. Extensive experiments on four fine-grained image retrieval benchmarks demonstrate that ConceptHash outperforms previous methods by a significant margin, offering unique sub-code interpretability as an additional benefit. Code at: https://github.com/kamwoh/concepthash., Comment: CVPRW 2024 - FGVC11 best paper award

The escalating frequency and intensity of heat-related climate events, particularly heatwaves, emphasize the pressing need for advanced heat risk management strategies. Current approaches, primarily relying on numerical models, face challenges in spatial-temporal resolution and in capturing the dynamic interplay of environmental, social, and behavioral factors affecting heat risks. This has led to difficulties in translating risk assessments into effective mitigation actions. Recognizing these problems, we introduce a novel approach leveraging the burgeoning capabilities of Large Language Models (LLMs) to extract rich and contextual insights from news reports. We hence propose an LLM-empowered visual analytics system, Havior, that integrates the precise, data-driven insights of numerical models with nuanced news report information. This hybrid approach enables a more comprehensive assessment of heat risks and better identification, assessment, and mitigation of heat-related threats. The system incorporates novel visualization designs, such as "thermoglyph" and news glyph, enhancing intuitive understanding and analysis of heat risks. The integration of LLM-based techniques also enables advanced information retrieval and semantic knowledge extraction that can be guided by experts' analytics needs. Our case studies on two cities that faced significant heatwave events and interviews with five experts have demonstrated the usefulness of our system in providing in-depth and actionable insights for heat risk management.

A seamless integration of neural networks with Isogeometric Analysis (IGA) was first introduced in [1] under the name of Hierarchical Deep-learning Neural Network (HiDeNN) and has systematically evolved into Isogeometric Convolution HiDeNN (in short, C-IGA) [2]. C-IGA achieves higher order approximations without increasing the degree of freedom. Due to the Kronecker delta property of C-IGA shape functions, one can refine the mesh in the physical domain like standard finite element method (FEM) while maintaining the exact geometrical mapping of IGA. In this article, C-IGA theory is generalized for multi-CAD-patch systems with a mathematical investigation of the compatibility conditions at patch interfaces and convergence of error estimates. Two compatibility conditions (nodal compatibility and G^0 (i.e., global C^0) compatibility) are presented and validated through numerical examples., Comment: 30 pages, 15 figures in main text, additional 10 pages for appendix

In this paper, we study the moments of the Age of Information (AoI) for both active and passive users in a random access network. In this network, active users broadcast sensing data, while passive users detect in-band radio activities from out-of-network devices, such as jammers. Collisions occur when multiple active users transmit simultaneously. Passive users can detect radio activities only when no active user transmits. Each active user's transmission behavior follows a Markov process. We aim to minimize the weighted sum of any moments of AoI for both user types. To achieve this, we employ a second-order analysis of system behavior. Specifically, we characterize an active user's transmission Markov process using its mean and temporal variance. We show that any moment of the AoI can be approximated by a function of these two parameters. This insight enables us to analyze and optimize the transmission Markov process for active users. We apply this strategy to two different random access models. Simulation results show that policies derived from this strategy outperform other baseline policies., Comment: Accepted by IEEE Transaction on Communications. arXiv admin note: text overlap with arXiv:2305.05137

The study of kagome materials has attracted much attention in the past few years due to the presence of many electron-electron interaction-driven phases in a single material.In this work, we report the discovery of intrinsic spin-polarized p-wave superconductivity in the thin-flake kagome material RbV$_3$Sb$_5$. Firstly, when an in-plane magnetic field is swept in opposite directions, we observe a unique form of hysteresis in magnetoresistance which is different from the hysteresis induced by extrinsic mechanisms such as flux-trapping or superheating and supercooling effects. The unconventional hysteresis indicates the emergence of an intrinsic time-reversal symmetry-breaking superconducting phase. Strikingly, at a fixed magnetic field, the finite-resistance state can be quenched to the zero-resistance state by applying a large current. Secondly, at temperatures around 400 mK, the re-entrance of superconductivity occurs during an in-plane field-sweeping process with a fixed sweeping direction. This kind of re-entrance is asymmetric about the zero field axis and observed in all field directions for a fixed current direction, which is different from the re-entrance observed in conventional superconductors. Moreover, the angle-dependent in-plane critical field measurements reveal a two-fold symmetry that deviates from the original, centrosymmetric D$_{6h}$ point group symmetry of the crystal. These findings put very strong constraints on the possible superconducting pairing symmetry of RbV$_3$Sb$_5$. We point out that the pairing symmetry, which is consistent with the crystal symmetry and all the observed novel properties, is a time-reversal symmetry-breaking, p-wave pairing with net spin polarization. Importantly, this p-wave pairing gives rise to a nodal topological superconducting state with Majorana flat bands on the sample edges., Comment: 21 pages, 4 figures

The Simons Observatory (SO) is a cosmic microwave background (CMB) survey experiment that includes small-aperture telescopes (SATs) observing from an altitude of 5,200 m in the Atacama Desert in Chile. The SO SATs will cover six spectral bands between 27 and 280 GHz to search for primordial B-modes to a sensitivity of $\sigma(r)=0.002$, with quantified systematic errors well below this value. Each SAT is a self-contained cryogenic telescope with a 35$^\circ$ field of view, 42 cm diameter optical aperture, 40 K half-wave plate, 1 K refractive optics, and $<0.1$ K focal plane that holds $>12,000$ TES detectors. We describe the nominal design of the SATs and present details about the integration and testing for one operating at 93 and 145 GHz.