Search

Learning world models offers a promising avenue for goal-conditioned reinforcement learning with sparse rewards. By allowing agents to plan actions or exploratory goals without direct interaction with the environment, world models enhance exploration efficiency. The quality of a world model hinges on the richness of data stored in the agent's replay buffer, with expectations of reasonable generalization across the state space surrounding recorded trajectories. However, challenges arise in generalizing learned world models to state transitions backward along recorded trajectories or between states across different trajectories, hindering their ability to accurately model real-world dynamics. To address these challenges, we introduce a novel goal-directed exploration algorithm, MUN (short for "World Models for Unconstrained Goal Navigation"). This algorithm is capable of modeling state transitions between arbitrary subgoal states in the replay buffer, thereby facilitating the learning of policies to navigate between any "key" states. Experimental results demonstrate that MUN strengthens the reliability of world models and significantly improves the policy's capacity to generalize across new goal settings., Comment: NeurIPS2024 Poster. arXiv admin note: substantial text overlap with arXiv:2411.01396

The Water Cherenkov Detector Array (WCDA) is one of the components of Large High Altitude Air Shower Observatory (LHAASO) and can monitor any sources over two-thirds of the sky for up to 7 hours per day with >98\% duty cycle. In this work, we report the detection of two outbursts of the Fanaroff-Riley I radio galaxy NGC 1275 that were detected by LHAASO-WCDA between November 2022 and January 2023 with statistical significance of 5.2~$\sigma$ and 8.3~$\sigma$. The observed spectral energy distribution in the range from 500 GeV to 3 TeV is fitted by a power-law with a best-fit spectral index of $\alpha=-3.37\pm0.52$ and $-3.35\pm0.29$, respectively. The outburst flux above 0.5~TeV was ($4.55\pm 4.21)\times~10^{-11}~\rm cm^{-2}~s^{-1}$ and ($3.45\pm 1.78)\times~10^{-11}~\rm cm^{-2}~s^{-1}$, corresponding to 60\%, 45\% of Crab Nebula flux. Variation analysis reveals the variability time-scale of days at the TeV energy band. A simple test by one-zone synchrotron self-Compton model reproduces the data in the gamma-ray band well., Comment: 11 pages, 8 figures, 3 tables

The rise of 6G-enable Vehicular Metaverses is transforming the automotive industry by integrating immersive, real-time vehicular services through ultra-low latency and high bandwidth connectivity. In 6G-enable Vehicular Metaverses, vehicles are represented by Vehicle Twins (VTs), which serve as digital replicas of physical vehicles to support real-time vehicular applications such as large Artificial Intelligence (AI) model-based Augmented Reality (AR) navigation, called VT tasks. VT tasks are resource-intensive and need to be offloaded to ground Base Stations (BSs) for fast processing. However, high demand for VT tasks and limited resources of ground BSs, pose significant resource allocation challenges, particularly in densely populated urban areas like intersections. As a promising solution, Unmanned Aerial Vehicles (UAVs) act as aerial edge servers to dynamically assist ground BSs in handling VT tasks, relieving resource pressure on ground BSs. However, due to high mobility of UAVs, there exists information asymmetry regarding VT task demands between UAVs and ground BSs, resulting in inefficient resource allocation of UAVs. To address these challenges, we propose a learning-based Modified Second-Bid (MSB) auction mechanism to optimize resource allocation between ground BSs and UAVs by accounting for VT task latency and accuracy. Moreover, we design a diffusion-based reinforcement learning algorithm to optimize the price scaling factor, maximizing the total surplus of resource providers and minimizing VT task latency. Finally, simulation results demonstrate that the proposed diffusion-based MSB auction outperforms traditional baselines, providing better resource distribution and enhanced service quality for vehicular users.

Recently, 3D Gaussian Splatting (3DGS) has revolutionized radiance field reconstruction, manifesting efficient and high-fidelity novel view synthesis. However, accurately representing surfaces, especially in large and complex scenarios, remains a significant challenge due to the unstructured nature of 3DGS. In this paper, we present CityGaussianV2, a novel approach for large-scale scene reconstruction that addresses critical challenges related to geometric accuracy and efficiency. Building on the favorable generalization capabilities of 2D Gaussian Splatting (2DGS), we address its convergence and scalability issues. Specifically, we implement a decomposed-gradient-based densification and depth regression technique to eliminate blurry artifacts and accelerate convergence. To scale up, we introduce an elongation filter that mitigates Gaussian count explosion caused by 2DGS degeneration. Furthermore, we optimize the CityGaussian pipeline for parallel training, achieving up to 10$\times$ compression, at least 25% savings in training time, and a 50% decrease in memory usage. We also established standard geometry benchmarks under large-scale scenes. Experimental results demonstrate that our method strikes a promising balance between visual quality, geometric accuracy, as well as storage and training costs. The project page is available at https://dekuliutesla.github.io/CityGaussianV2/., Comment: Project Page: https://dekuliutesla.github.io/CityGaussianV2/

Recent advances in AI-generated voices have intensified the challenge of detecting deepfake audio, posing risks for scams and the spread of disinformation. To tackle this issue, we establish the largest public voice dataset to date, named DeepFakeVox-HQ, comprising 1.3 million samples, including 270,000 high-quality deepfake samples from 14 diverse sources. Despite previously reported high accuracy, existing deepfake voice detectors struggle with our diversely collected dataset, and their detection success rates drop even further under realistic corruptions and adversarial attacks. We conduct a holistic investigation into factors that enhance model robustness and show that incorporating a diversified set of voice augmentations is beneficial. Moreover, we find that the best detection models often rely on high-frequency features, which are imperceptible to humans and can be easily manipulated by an attacker. To address this, we propose the F-SAT: Frequency-Selective Adversarial Training method focusing on high-frequency components. Empirical results demonstrate that using our training dataset boosts baseline model performance (without robust training) by 33%, and our robust training further improves accuracy by 7.7% on clean samples and by 29.3% on corrupted and attacked samples, over the state-of-the-art RawNet3 model.

Using early data from the Dark Energy Spectroscopic Instrument (DESI) survey, we search for AGN signatures in 410,757 line-emitting galaxies. By employing the BPT emission-line ratio diagnostic diagram, we identify AGN in 75,928/296,261 ($\approx$25.6%) high-mass ($\log (M_{\star}/\rm M_{\odot}) >$ 9.5) and 2,444/114,496 ($\approx$2.1%) dwarf ($\log (M_{\star}/\rm M_{\odot}) \leq$ 9.5) galaxies. Of these AGN candidates, 4,181 sources exhibit a broad H$\alpha$ component, allowing us to estimate their BH masses via virial techniques. This study more than triples the census of dwarf AGN as well as that of intermediate-mass black hole (IMBH; $M_{\rm BH} \le 10^6~\rm M_{\odot}$) candidates, spanning a broad discovery space in stellar mass (7 $< \log (M_{\star}/\rm M_{\odot}) <$ 12) and redshift (0.001 $< \rm z <$ 0.45). The observed AGN fraction in dwarf galaxies ($\approx$2.1%) is nearly four times higher than prior estimates, primarily due to DESI's smaller fiber size, which enables the detection of lower luminosity dwarf AGN candidates. We also extend the $M_{\rm BH}$ - $M_{\star}$ scaling relation down to $\log (M_{\star}/\rm M_{\odot}) \approx$ 8.5 and $\log (M_{\rm BH}/M_{\odot}) \approx$ 4.4, with our results aligning well with previous low-redshift studies. The large statistical sample of dwarf AGN candidates from current and future DESI releases will be invaluable for enhancing our understanding of galaxy evolution at the low-mass end of the galaxy mass function., Comment: 35 pages, 22 figures, Submitted to AAS Journals, Comments are welcome

In autonomous robot navigation, terrain cost assignment is typically performed using a semantics-based paradigm in which terrain is first labeled using a pre-trained semantic classifier and costs are then assigned according to a user-defined mapping between label and cost. While this approach is rapidly adaptable to changing user preferences, only preferences over the types of terrain that are already known by the semantic classifier can be expressed. In this paper, we hypothesize that a machine-learning-based alternative to the semantics-based paradigm above will allow for rapid cost assignment adaptation to preferences expressed over new terrains at deployment time without the need for additional training. To investigate this hypothesis, we introduce and study PACER, a novel approach to costmap generation that accepts as input a single birds-eye view (BEV) image of the surrounding area along with a user-specified preference context and generates a corresponding BEV costmap that aligns with the preference context. Using both real and synthetic data along with a combination of proposed training tasks, we find that PACER is able to adapt quickly to new user preferences while also exhibiting better generalization to novel terrains compared to both semantics-based and representation-learning approaches.

Generalization to novel object configurations and instances across diverse tasks and environments is a critical challenge in robotics. Keypoint-based representations have been proven effective as a succinct representation for capturing essential object features, and for establishing a reference frame in action prediction, enabling data-efficient learning of robot skills. However, their manual design nature and reliance on additional human labels limit their scalability. In this paper, we propose KALM, a framework that leverages large pre-trained vision-language models (LMs) to automatically generate task-relevant and cross-instance consistent keypoints. KALM distills robust and consistent keypoints across views and objects by generating proposals using LMs and verifies them against a small set of robot demonstration data. Based on the generated keypoints, we can train keypoint-conditioned policy models that predict actions in keypoint-centric frames, enabling robots to generalize effectively across varying object poses, camera views, and object instances with similar functional shapes. Our method demonstrates strong performance in the real world, adapting to different tasks and environments from only a handful of demonstrations while requiring no additional labels. Website: https://kalm-il.github.io/, Comment: CoRL LangRob Workshop, 2024

Deep visual Simultaneous Localization and Mapping (SLAM) techniques, e.g., DROID, have made significant advancements by leveraging deep visual odometry on dense flow fields. In general, they heavily rely on global visual similarity matching. However, the ambiguous similarity interference in uncertain regions could often lead to excessive noise in correspondences, ultimately misleading SLAM in geometric modeling. To address this issue, we propose a Learnable Gaussian Uncertainty (LGU) matching. It mainly focuses on precise correspondence construction. In our scheme, a learnable 2D Gaussian uncertainty model is designed to associate matching-frame pairs. It could generate input-dependent Gaussian distributions for each correspondence map. Additionally, a multi-scale deformable correlation sampling strategy is devised to adaptively fine-tune the sampling of each direction by a priori look-up ranges, enabling reliable correlation construction. Furthermore, a KAN-bias GRU component is adopted to improve a temporal iterative enhancement for accomplishing sophisticated spatio-temporal modeling with limited parameters. The extensive experiments on real-world and synthetic datasets are conducted to validate the effectiveness and superiority of our method.

Retrieval-Augmented Generation (RAG) has become a powerful paradigm for enhancing large language models (LLMs) through external knowledge retrieval. Despite its widespread attention, existing academic research predominantly focuses on single-turn RAG, leaving a significant gap in addressing the complexities of multi-turn conversations found in real-world applications. To bridge this gap, we introduce CORAL, a large-scale benchmark designed to assess RAG systems in realistic multi-turn conversational settings. CORAL includes diverse information-seeking conversations automatically derived from Wikipedia and tackles key challenges such as open-domain coverage, knowledge intensity, free-form responses, and topic shifts. It supports three core tasks of conversational RAG: passage retrieval, response generation, and citation labeling. We propose a unified framework to standardize various conversational RAG methods and conduct a comprehensive evaluation of these methods on CORAL, demonstrating substantial opportunities for improving existing approaches.

Multivariate time series forecasting plays a crucial role in various real-world applications. Significant efforts have been made to integrate advanced network architectures and training strategies that enhance the capture of temporal dependencies, thereby improving forecasting accuracy. On the other hand, mainstream approaches typically utilize a single unified model with simplistic channel-mixing embedding or cross-channel attention operations to account for the critical intricate inter-channel dependencies. Moreover, some methods even trade capacity for robust prediction based on the channel-independent assumption. Nonetheless, as time series data may display distinct evolving patterns due to the unique characteristics of each channel (including multiple strong seasonalities and trend changes), the unified modeling methods could yield suboptimal results. To this end, we propose DisenTS, a tailored framework for modeling disentangled channel evolving patterns in general multivariate time series forecasting. The central idea of DisenTS is to model the potential diverse patterns within the multivariate time series data in a decoupled manner. Technically, the framework employs multiple distinct forecasting models, each tasked with uncovering a unique evolving pattern. To guide the learning process without supervision of pattern partition, we introduce a novel Forecaster Aware Gate (FAG) module that generates the routing signals adaptively according to both the forecasters' states and input series' characteristics. The forecasters' states are derived from the Linear Weight Approximation (LWA) strategy, which quantizes the complex deep neural networks into compact matrices. Additionally, the Similarity Constraint (SC) is further proposed to guide each model to specialize in an underlying pattern by minimizing the mutual information between the representations.

The application of reinforcement learning in traffic signal control (TSC) has been extensively researched and yielded notable achievements. However, most existing works for TSC assume that traffic data from all surrounding intersections is fully and continuously available through sensors. In real-world applications, this assumption often fails due to sensor malfunctions or data loss, making TSC with missing data a critical challenge. To meet the needs of practical applications, we introduce DiffLight, a novel conditional diffusion model for TSC under data-missing scenarios in the offline setting. Specifically, we integrate two essential sub-tasks, i.e., traffic data imputation and decision-making, by leveraging a Partial Rewards Conditioned Diffusion (PRCD) model to prevent missing rewards from interfering with the learning process. Meanwhile, to effectively capture the spatial-temporal dependencies among intersections, we design a Spatial-Temporal transFormer (STFormer) architecture. In addition, we propose a Diffusion Communication Mechanism (DCM) to promote better communication and control performance under data-missing scenarios. Extensive experiments on five datasets with various data-missing scenarios demonstrate that DiffLight is an effective controller to address TSC with missing data. The code of DiffLight is released at https://github.com/lokol5579/DiffLight-release., Comment: Accepted by NeurIPS 2024

In this work, we introduce novel information-theoretic generalization bounds using the conditional $f$-information framework, an extension of the traditional conditional mutual information (MI) framework. We provide a generic approach to derive generalization bounds via $f$-information in the supersample setting, applicable to both bounded and unbounded loss functions. Unlike previous MI-based bounds, our proof strategy does not rely on upper bounding the cumulant-generating function (CGF) in the variational formula of MI. Instead, we set the CGF or its upper bound to zero by carefully selecting the measurable function invoked in the variational formula. Although some of our techniques are partially inspired by recent advances in the coin-betting framework (e.g., Jang et al. (2023)), our results are independent of any previous findings from regret guarantees of online gambling algorithms. Additionally, our newly derived MI-based bound recovers many previous results and improves our understanding of their potential limitations. Finally, we empirically compare various $f$-information measures for generalization, demonstrating the improvement of our new bounds over the previous bounds., Comment: Accepted at NeurIPS 2024

The simultaneous presence of polarity and metallicity in a material signifies an exotic polar metal state, but such materials are extremely rare, especially in bulk form, due to mutually exclusive nature of the fundamental defining properties. Here, we report experimental findings that HgPSe3 is a robust bulk polar metal at room temperature with a chiral structure stabilized by pressure and, remarkably, this polar metal hosts superconductivity with critical temperature Tc up to 11 K. Theoretical analysis reveals a two-step interlayer sliding-then-compressing mechanism for coexistence of polarity and metallicity in HgPSe3. This work unveils a new paradigm for creating the bulk polar metal state and simultaneous presence of coexisting quantum orders, raising the prospect of discovering novel emergent physics using pressure as a tuning knob., Comment: 19 pages, 4 main figures + 6 extented figures

We theoretically propose a mechanism to realize the superconducting diode effect (SDE): The current can generate a magnetic field, affecting the magnetic moment of magnetic impurity. When the connection region of the Josephson junction is coupled with the magnetic impurity, the supercurrents in positive and negative directions have different influences on the magnetic moment. This results in a phenomenon that the critical supercurrents in these opposite directions are unequal, which is called SDE. We model the Josephson connection region by a quantum dot. Then the critical supercurrents are investigated by the non-equilibrium Green's function method, and we carry out a detailed symmetry analysis on the supercurrent relations. The calculation results confirm that the SDE does exist in this system. Besides, the SDE is significant in a wide parameter space and can be effectively adjusted in various ways. Our design only demands a magnetic impurity and conventional superconductors. The unconventional finite-momentum Cooper pair and spin-orbit coupling are not required, and there is also no need for the existence of chirality or an external magnetic field. Our work provides a universal device structure for the development of superconducting electronics., Comment: 10 pages, 6 figures

We propose a superconducting phase-controlled thermal switch based on a four-terminal graphene-superconductor system. By the coupling of two superconducting leads on a zigzag graphene nanoribbon, both the normal-transmission coefficient and the crossed-Andreev-reflection coefficient, which dominate the thermal conductivity of electrons in the graphene nanoribbon, can be well controlled simultaneously by the phase difference of the superconducting leads. As a result, the thermal conductivity of electrons in the graphene nanoribbon can be tuned and a thermal switching effect appears. Using the nonequilibrium Green's function method, we verify this thermal switching effect numerically. At ambient temperatures less than about one tenth of the superconducting transition temperature, the thermal switching ratio can exceed 2000. The performance of the thermal switch can be regulated by the ambient temperature, and doping or gating can slightly increase the thermal switching ratio. The use of narrower graphene nanoribbons and wider superconducting leads facilitates the obtaining of larger thermal switching ratios. This switching effect of electronic thermal conductance in graphene is expected to be experimentally realized and applied., Comment: 9 pages, 5 figures

Generating overtaking trajectories in high-speed scenarios presents significant challenges and is typically addressed through hierarchical planning methods. However, this method has two primary drawbacks. First, heuristic algorithms can only provide a single initial solution, which may lead to local optima and consequently diminish the quality of the solution. Second, the time efficiency of trajectory refinement based on numerical optimization is insufficient. To overcome these limitations, this paper proposes an overtaking trajectory planning framework based on spatio-temporal topology and reachable set analysis (SROP), to improve trajectory quality and time efficiency. Specifically, this paper introduces topological classes to describe trajectories representing different overtaking behaviors, which support the spatio-temporal topological search method employed by the upper-layer planner to identify diverse initial paths. This approach helps prevent getting stuck in local optima, enhancing the overall solution quality by considering multiple initial solutions from distinct topologies. Moreover, the reachable set method is integrated into the lower-layer planner for parallel trajectory evaluation. This method enhances planning efficiency by decoupling vehicle model constraints from the optimization process, enabling parallel computation while ensuring control feasibility. Simulation results show that the proposed method improves the smoothness of generated trajectories by 66.8% compared to state-of-the-art methods, highlighting its effectiveness in enhancing trajectory quality. Additionally, this method reduces computation time by 62.9%, demonstrating its efficiency.

Exceptional points (EPs) have been extensively explored in mechanical, acoustic, plasmonic, and photonic systems. However, little is known about the role of EPs in tailoring the dynamic tunability of optical devices. A specific type of EPs known as chiral EPs has recently attracted much attention for controlling the flow of light and for building sensors with better responsivity. A recently demonstrated route to chiral EPs via lithographically defined symmetric Mie scatterers on the rim of resonators has not only provided the much-needed mechanical stability for studying chiral EPs but also helped reduce losses originating from nanofabrication imperfections, facilitating the in-situ study of chiral EPs and their contribution to the dynamics and tunability of resonators. Here, we use asymmetric Mie scatterers to break the rotational symmetry of a microresonator, to demonstrate deterministic thermal tuning across a chiral EP, and to demonstrate EP-mediated chiral optical nonlinear response and efficient electro-optic tuning. Our results indicate asymmetric electro-optic modulation with up to 17dB contrast at GHz and CMOS-compatible voltage levels. Such wafer-scale nano-manufacturing of chiral electro-optic modulators and the chiral EP-tailored tunning may facilitate new micro-resonator functionalities in quantum information processing, electromagnetic wave control, and optical interconnects.

The study of cells and their responses to genetic or chemical perturbations promises to accelerate the discovery of therapeutic targets. However, designing adequate and insightful models for such data is difficult because the response of a cell to perturbations essentially depends on its biological context (e.g., genetic background or cell type). For example, while discovering therapeutic targets, one may want to enrich for drugs that specifically target a certain cell type. This challenge emphasizes the need for methods that explicitly take into account potential interactions between drugs and contexts. Towards this goal, we propose a novel Factorized Causal Representation (FCR) learning method that reveals causal structure in single-cell perturbation data from several cell lines. Based on the framework of identifiable deep generative models, FCR learns multiple cellular representations that are disentangled, comprised of covariate-specific ($\mathbf{z}_x$), treatment-specific ($\mathbf{z}_{t}$), and interaction-specific ($\mathbf{z}_{tx}$) blocks. Based on recent advances in non-linear ICA theory, we prove the component-wise identifiability of $\mathbf{z}_{tx}$ and block-wise identifiability of $\mathbf{z}_t$ and $\mathbf{z}_x$. Then, we present our implementation of FCR, and empirically demonstrate that it outperforms state-of-the-art baselines in various tasks across four single-cell datasets.

Online shopping is a complex multi-task, few-shot learning problem with a wide and evolving range of entities, relations, and tasks. However, existing models and benchmarks are commonly tailored to specific tasks, falling short of capturing the full complexity of online shopping. Large Language Models (LLMs), with their multi-task and few-shot learning abilities, have the potential to profoundly transform online shopping by alleviating task-specific engineering efforts and by providing users with interactive conversations. Despite the potential, LLMs face unique challenges in online shopping, such as domain-specific concepts, implicit knowledge, and heterogeneous user behaviors. Motivated by the potential and challenges, we propose Shopping MMLU, a diverse multi-task online shopping benchmark derived from real-world Amazon data. Shopping MMLU consists of 57 tasks covering 4 major shopping skills: concept understanding, knowledge reasoning, user behavior alignment, and multi-linguality, and can thus comprehensively evaluate the abilities of LLMs as general shop assistants. With Shopping MMLU, we benchmark over 20 existing LLMs and uncover valuable insights about practices and prospects of building versatile LLM-based shop assistants. Shopping MMLU can be publicly accessed at https://github.com/KL4805/ShoppingMMLU. In addition, with Shopping MMLU, we host a competition in KDD Cup 2024 with over 500 participating teams. The winning solutions and the associated workshop can be accessed at our website https://amazon-kddcup24.github.io/., Comment: NeurIPS 2024 Datasets and Benchmarks Track Accepted. Modified typos in Figure 9

Without applied bias voltage, a superconducting phase difference can drive a charge Josephson supercurrent in a superconductor junction. In analogy, we here theoretically propose a spin phase that intrinsically generates spin Josephson supercurrent, and this spin Josephson effect is studied in a junction of superconducting nanowire (SNW). We show that spin-orbit coupling and magnetic field give rise to spin-triplet superconductivity in the SNW, thus allow the superfluid of both charge and spin. Next we introduce the concept of spin phase that can be generated by controls of spin current, magnetic field or electric field. By the analysis of pairing correlations and Ginzburg-Landau-type theory, it is shown that the spin phase makes spin-up and spin-down $S=1$ Cooper pairs get opposite phases and move oppositely in a Josephson junction, so that a dissipationless pure spin current is induced. We also derive the formula that the spin current is equal to the derivative of Andreev bound state to the spin phase, which is analogous to that of charge Josephson effect. At last, our calculation verifies the existence of spin supercurrent inside the superconducting gap in the topologically nontrivial phase. Our study provides a view on spin Josephson effect and indicates the potential combination of topological superconductivity and spintronics., Comment: 9 pages, 3 figures

Zero-bias charge conductance peak (ZBCCP) is a significant symbol of Majorana zero modes (MZMs). The proximity effect of s-wave superconductor is usually demanded in the fabrication of MZMs. So in transport experiments, the system is inevitably coupled to the s-wave superconductor. Here we study how the ZBCCP is affected by coupling of the s-wave superconductor. The results show that the ZBCCP could be changed into a zero-bias valley due to the coupling of s-wave superconductor, although the conductance at the zero bias still keeps a quantized value $2e^2/h$. So it does not mean no MZM exists when no ZBCCP is experimentally observed. In addition, the spin transport is investigated. Four reflection processes (the normal reflection, spin-flip reflection, normal Andreev reflection, and equal-spin Andreev reflection) usually occur. The reflection coefficients are strongly dependent on the spin direction of the incident electron, and they may be symmetrical or Fano resonance shapes. But the spin conductance always shows a zero-bias peak with the height $e/2\pi$ regardless of the direction of the spin bias and the coupling strength of the s-wave superconductor. So measuring spin transport properties could be a more reliable method to judge the existence of MZMs., Comment: 8 pages, 5 figures

GPRec explicitly categorizes users into groups in a learnable manner and aligns them with corresponding group embeddings. We design the dual group embedding space to offer a diverse perspective on group preferences by contrasting positive and negative patterns. On the individual level, GPRec identifies personal preferences from ID-like features and refines the obtained individual representations to be independent of group ones, thereby providing a robust complement to the group-level modeling. We also present various strategies for the flexible integration of GPRec into various DRS models. Rigorous testing of GPRec on three public datasets has demonstrated significant improvements in recommendation quality.

The combination of spin-orbit coupling and superconductivity induces unconventional spin-triplet correlation in Ising superconductors. We theoretically investigate the spin transport through a normal metal-Ising superconductor junction, showing that Ising superconductors also have the characteristic of spin superconductivity.Due to the existence of spin-triplet Cooper pairs, not only charge supercurrent but also spin supercurrent can transport in Ising superconductors.We analyze the transport process in the junction which is mainly contributed by the equal-spin Andreev reflection and spin-flip reflection, and calculate the spin conductance and the spin injection efficiency under different conditions.Our findings broaden the boundary of spin superconductivity and reveal the potential applications of Ising superconductors in spintronics, especially in controlled long-distance dissipationless spin transport., Comment: 8 pages,6 figures

In public safety and social life, the task of Clothes-Changing Person Re-Identification (CC-ReID) has become increasingly significant. However, this task faces considerable challenges due to appearance changes caused by clothing alterations. Addressing this issue, this paper proposes an innovative method for disentangled feature extraction, effectively extracting discriminative features from pedestrian images that are invariant to clothing. This method leverages pedestrian parsing techniques to identify and retain features closely associated with individual identity while disregarding the variable nature of clothing attributes. Furthermore, this study introduces a gated channel attention mechanism, which, by adjusting the network's focus, aids the model in more effectively learning and emphasizing features critical for pedestrian identity recognition. Extensive experiments conducted on two standard CC-ReID datasets validate the effectiveness of the proposed approach, with performance surpassing current leading solutions. The Top-1 accuracy under clothing change scenarios on the PRCC and VC-Clothes datasets reached 64.8% and 83.7%, respectively., Comment: The article has been accepted by IEEE International Conference on Multimedia and Expo 2024

We report the discovery of a peculiar X-ray transient, EP240408a, by Einstein Probe (EP) and follow-up studies made with EP, Swift, NICER, GROND, ATCA and other ground-based multi-wavelength telescopes. The new transient was first detected with Wide-field X-ray Telescope (WXT) on board EP on April 8th, 2024, manifested in an intense yet brief X-ray flare lasting for 12 seconds. The flare reached a peak flux of 3.9x10^(-9) erg/cm2/s in 0.5-4 keV, about 300 times brighter than the underlying X-ray emission detected throughout the observation. Rapid and more precise follow-up observations by EP/FXT, Swift and NICER confirmed the finding of this new transient. Its X-ray spectrum is non-thermal in 0.5-10 keV, with a power-law photon index varying within 1.8-2.5. The X-ray light curve shows a plateau lasting for about 4 days, followed by a steep decay till becoming undetectable about 10 days after the initial detection. Based on its temporal property and constraints from previous EP observations, an unusual timescale in the range of 7-23 days is found for EP240408a, which is intermediate between the commonly found fast and long-term transients. No counterparts have been found in optical and near-infrared, with the earliest observation at 17 hours after the initial X-ray detection, suggestive of intrinsically weak emission in these bands. We demonstrate that the remarkable properties of EP240408a are inconsistent with any of the transient types known so far, by comparison with, in particular, jetted tidal disruption events, gamma-ray bursts, X-ray binaries and fast blue optical transients. The nature of EP240408a thus remains an enigma. We suggest that EP240408a may represent a new type of transients with intermediate timescales of the order of about 10 days. The detection and follow-ups of more of such objects are essential for revealing their origin., Comment: 25 pages, 11 figures

Large language models (LLMs) have been widely deployed as the backbone with additional tools and text information for real-world applications. However, integrating external information into LLM-integrated applications raises significant security concerns. Among these, prompt injection attacks are particularly threatening, where malicious instructions injected in the external text information can exploit LLMs to generate answers as the attackers desire. While both training-time and test-time defense methods have been developed to mitigate such attacks, the unaffordable training costs associated with training-time methods and the limited effectiveness of existing test-time methods make them impractical. This paper introduces a novel test-time defense strategy, named Formatting AuThentication with Hash-based tags (FATH). Unlike existing approaches that prevent LLMs from answering additional instructions in external text, our method implements an authentication system, requiring LLMs to answer all received instructions with a security policy and selectively filter out responses to user instructions as the final output. To achieve this, we utilize hash-based authentication tags to label each response, facilitating accurate identification of responses according to the user's instructions and improving the robustness against adaptive attacks. Comprehensive experiments demonstrate that our defense method can effectively defend against indirect prompt injection attacks, achieving state-of-the-art performance under Llama3 and GPT3.5 models across various attack methods. Our code is released at: https://github.com/Jayfeather1024/FATH

By pretraining on trillions of tokens, an LLM gains the capability of text generation. However, to enhance its utility and reduce potential harm, SFT and alignment are applied sequentially to the pretrained model. Due to the differing nature and objective functions of SFT and alignment, catastrophic forgetting has become a significant issue. To address this, we introduce Unified Fine-Tuning (UFT), which integrates SFT and alignment into a single training stage using the same objective and loss functions through an implicit reward function. Our experimental results demonstrate that UFT outperforms SFT on instruction-tuning data alone. Moreover, when combining instruction-tuning data with alignment data, UFT effectively prevents catastrophic forgetting across these two stages and shows a clear advantage over sequentially applying SFT and alignment. This is evident in the significant improvements observed in the \textbf{ifeval} task for instruction-following and the \textbf{truthful-qa} task for factuality. The proposed general fine-tuning framework UFT establishes an effective and efficient pretraining-UFT paradigm for LLM training.

A wide variety of data can be represented using third-order tensors, spanning applications in chemometrics, psychometrics, and image processing. However, traditional data-driven frameworks are not naturally equipped to process tensors without first unfolding or flattening the data, which can result in a loss of crucial higher-order structural information. In this article, we introduce a novel framework for the data-driven analysis of T-product-based dynamical systems (TPDSs), where the system evolution is governed by the T-product between a third-order dynamic tensor and a third-order state tensor. In particular, we examine the data informativity of TPDSs concerning system identification, stability, controllability, and stabilizability and illustrate significant computational improvements over traditional approaches by leveraging the unique properties of the T-product. The effectiveness of our framework is demonstrated through numerical examples., Comment: 11 pages, 3 figures

A major limitation of machine learning (ML) prediction models is that they recover associational, rather than causal, predictive relationships between variables. In high-stakes automation applications of ML this is problematic, as the model often learns spurious, non-causal associations. This paper proposes mechanism learning, a simple method which uses front-door causal bootstrapping to deconfound observational data such that any appropriate ML model is forced to learn predictive relationships between effects and their causes (reverse causal inference), despite the potential presence of multiple unknown and unmeasured confounding. Effect variables can be very high dimensional, and the predictive relationship nonlinear, as is common in ML applications. This novel method is widely applicable, the only requirement is the existence of a mechanism variable mediating the cause (prediction target) and effect (feature data), which is independent of the (unmeasured) confounding variables. We test our method on fully synthetic, semi-synthetic and real-world datasets, demonstrating that it can discover reliable, unbiased, causal ML predictors where by contrast, the same ML predictor trained naively using classical supervised learning on the original observational data, is heavily biased by spurious associations. We provide code to implement the results in the paper, online., Comment: 12 pages, 6 figures

In offline reinforcement learning (RL), addressing the out-of-distribution (OOD) action issue has been a focus, but we argue that there exists an OOD state issue that also impairs performance yet has been underexplored. Such an issue describes the scenario when the agent encounters states out of the offline dataset during the test phase, leading to uncontrolled behavior and performance degradation. To this end, we propose SCAS, a simple yet effective approach that unifies OOD state correction and OOD action suppression in offline RL. Technically, SCAS achieves value-aware OOD state correction, capable of correcting the agent from OOD states to high-value in-distribution states. Theoretical and empirical results show that SCAS also exhibits the effect of suppressing OOD actions. On standard offline RL benchmarks, SCAS achieves excellent performance without additional hyperparameter tuning. Moreover, benefiting from its OOD state correction feature, SCAS demonstrates enhanced robustness against environmental perturbations., Comment: Accepted to NeurIPS 2024

In scenarios with limited training data or where explainability is crucial, conventional neural network-based machine learning models often face challenges. In contrast, Bayesian inference-based algorithms excel in providing interpretable predictions and reliable uncertainty estimation in these scenarios. While many state-of-the-art in-memory computing (IMC) architectures leverage emerging non-volatile memory (NVM) technologies to offer unparalleled computing capacity and energy efficiency for neural network workloads, their application in Bayesian inference is limited. This is because the core operations in Bayesian inference differ significantly from the multiplication-accumulation (MAC) operations common in neural networks, rendering them generally unsuitable for direct implementation in most existing IMC designs. In this paper, we propose FeBiM, an efficient and compact Bayesian inference engine powered by multi-bit ferroelectric field-effect transistor (FeFET)-based IMC. FeBiM effectively encodes the trained probabilities of a Bayesian inference model within a compact FeFET-based crossbar. It maps quantized logarithmic probabilities to discrete FeFET states. As a result, the accumulated outputs of the crossbar naturally represent the posterior probabilities, i.e., the Bayesian inference model's output given a set of observations. This approach enables efficient in-memory Bayesian inference without the need for additional calculation circuitry. As the first FeFET-based in-memory Bayesian inference engine, FeBiM achieves an impressive storage density of 26.32 Mb/mm$^{2}$ and a computing efficiency of 581.40 TOPS/W in a representative Bayesian classification task. These results demonstrate 10.7$\times$/43.4$\times$ improvement in compactness/efficiency compared to the state-of-the-art hardware implementation of Bayesian inference., Comment: 6 pages, 8 figures, to be published in the 61st DAC (Design Automation Conference) proceedings

Effective control of magnetic phases in two-dimensional magnets would constitute crucial progress in spintronics, holding great potential for future computing technologies. Here, we report a new approach of leveraging tunneling current as a tool for controlling spin states in CrI3. We reveal that a tunneling current can deterministically switch between spin-parallel and spin-antiparallel states in few-layer CrI3, depending on the polarity and amplitude of the current. We propose a mechanism involving nonequilibrium spin accumulation in the graphene electrodes in contact with the CrI3 layers. We further demonstrate tunneling current-tunable stochastic switching between multiple spin states of the CrI3 tunnel devices, which goes beyond conventional bi-stable stochastic magnetic tunnel junctions and has not been documented in two-dimensional magnets. Our findings not only address the existing knowledge gap concerning the influence of tunneling currents in controlling the magnetism in two-dimensional magnets, but also unlock possibilities for energy-efficient probabilistic and neuromorphic computing., Comment: 23 pages, 4 figures

Several medical Multimodal Large Languange Models (MLLMs) have been developed to address tasks involving visual images with textual instructions across various medical modalities, achieving impressive results. Most current medical generalist models are region-agnostic, treating the entire image as a holistic representation. However, they struggle to identify which specific regions they are focusing on when generating a sentence. To mimic the behavior of doctors, who typically begin by reviewing the entire image before concentrating on specific regions for a thorough evaluation, we aim to enhance the capability of medical MLLMs in understanding anatomical regions within entire medical scans. To achieve it, we first formulate Region-Centric tasks and construct a large-scale dataset, MedRegInstruct, to incorporate regional information into training. Combining our collected dataset with other medical multimodal corpora for training, we propose a Region-Aware medical MLLM, MedRegA, which is the first bilingual generalist medical AI system to simultaneously handle image-level and region-level medical vision-language tasks across a broad range of modalities. Our MedRegA not only enables three region-centric tasks, but also achieves the best performance for visual question answering, report generation and medical image classification over 8 modalities, showcasing significant versatility. Experiments demonstrate that our model can not only accomplish powerful performance across various medical vision-language tasks in bilingual settings, but also recognize and detect structures in multimodal medical scans, boosting the interpretability and user interactivity of medical MLLMs. Our project page is https://medrega.github.io., Comment: Technical Report

Factuality evaluation aims to detect factual errors produced by language models (LMs) and hence guide the development of more factual models. Towards this goal, we train a factuality evaluator, FenCE, that provides LM generators with claim-level factuality feedback. We conduct data augmentation on a combination of public judgment datasets to train FenCE to (1) generate textual critiques along with scores and (2) make claim-level judgment based on diverse source documents obtained by various tools. We then present a framework that leverages FenCE to improve the factuality of LM generators by constructing training data. Specifically, we generate a set of candidate responses, leverage FenCE to revise and score each response without introducing lesser-known facts, and train the generator by preferring highly scored revised responses. Experiments show that our data augmentation methods improve the evaluator's accuracy by 2.9% on LLM-AggreFact. With FenCE, we improve Llama3-8B-chat's factuality rate by 14.45% on FActScore, outperforming state-of-the-art factuality finetuning methods by 6.96%.

LEIA (Lobster Eye Imager for Astronomy) detected a new X-ray transient on November 7, 2022, identified as a superflare event occurring on a nearby RS CVn-type binary HD 251108. The flux increase was also detected in follow-up observations at X-ray, UV and optical wavelengths. The flare lasted for about 40 days in soft X-ray observations, reaching a peak luminosity of ~1.1 * 10^34 erg/s in 0.5-4.0 keV, which is roughly 60 times the quiescent luminosity. Optical brightening was observed for only one night. The X-ray light curve is well described by a double "FRED" (fast rise and exponential decay) model, attributed to the cooling process of a loop arcade structure formed subsequent to the initial large loop with a half-length of ~1.9 times the radius of the host star. Time-resolved X-ray spectra were fitted with a two-temperature apec model, showing significant evolution of plasma temperature, emission measure, and metal abundance over time. The estimated energy released in the LEIA band is ~3 * 10^39 erg, suggesting this is likely the most energetic X-ray stellar flare with the longest duration detected to date., Comment: submitted to ApJL, 22 pages, 9 figures, 7 tables

Hyperbolic neural networks (HNNs) have been proved effective in modeling complex data structures. However, previous works mainly focused on the Poincar\'e ball model and the hyperboloid model as coordinate representations of the hyperbolic space, often neglecting the Klein model. Despite this, the Klein model offers its distinct advantages thanks to its straight-line geodesics, which facilitates the well-known Einstein midpoint construction, previously leveraged to accompany HNNs in other models. In this work, we introduce a framework for hyperbolic neural networks based on the Klein model. We provide detailed formulation for representing useful operations using the Klein model. We further study the Klein linear layer and prove that the "tangent space construction" of the scalar multiplication and parallel transport are exactly the Einstein scalar multiplication and the Einstein addition, analogous to the M\"obius operations used in the Poincar\'e ball model. We show numerically that the Klein HNN performs on par with the Poincar\'e ball model, providing a third option for HNN that works as a building block for more complicated architectures., Comment: Accepted to NeurIPS 2024 Symmetry and Geometry in Neural Representations Workshop

We analyzed the global and resolved properties of approximately 1,240 nearby star-forming galaxies from the MaNGA survey, comparing compact and extended galaxies -- those with smaller and larger radii ($R_{\rm e}$), respectively -- at a fixed stellar mass ($M_{\ast}$). Compact galaxies typically exhibit lower HI gas fractions, higher dust extinction, higher metallicity, greater mass concentration, and lower angular momentum on a global scale. Radial profiles of stellar mass surface density ($\Sigma_{\ast}$) and star formation rate surface density ($\Sigma_{\rm SFR}$), as functions of the effective radius ($R/$$R_{\rm e}$), reveal that compact galaxies display steeper gradients and higher values, resulting in elevated specific star formation rates (sSFR) in their inner regions compared to their outskirts. At a given $\Sigma_{\ast}$, compact galaxies have higher sSFR than extended galaxies, particularly in low-mass galaxies (log($M_{\ast}$/$M_{\odot}$)$\,\leq\,$10$^{10}$). Additionally, their metallicity profiles differ significantly: extended galaxies have steeper metallicity gradients, while compact galaxies exhibit flatter slopes and higher metallicity at a given $R/$$R_{\rm e}$. After accounting for the dependence of metallicity on $M_{\ast}$ and $\Sigma_{\ast}$, no further correlation with SFR is observed. The combination of higher sSFR and potentially higher star formation efficiency in compact galaxies suggests that their central gas is being rapidly consumed, leading to older stellar populations, as indicated by D$_{n}$(4000) and EW(H$\delta_A$), and resulting in faster central growth. Our results reveal that radial SFR profiles cannot be fully determined by $M_{\ast}$ and $\Sigma_{\ast}$ alone; other factors, such as galaxy size or angular momentum, must be considered to fully understand the observed trends., Comment: 22 pages, 12 figures, ApJ accepted

Deep learning techniques are dominating automated animal activity recognition (AAR) tasks with wearable sensors due to their high performance on large-scale labelled data. However, current deep learning-based AAR models are trained solely on datasets of individual animal species, constraining their applicability in practice and performing poorly when training data are limited. In this study, we propose a one-for-many framework, dubbed Cross-species Knowledge Sharing and Preserving (CKSP), based on sensor data of diverse animal species. Given the coexistence of generic and species-specific behavioural patterns among different species, we design a Shared-Preserved Convolution (SPConv) module. This module assigns an individual low-rank convolutional layer to each species for extracting species-specific features and employs a shared full-rank convolutional layer to learn generic features, enabling the CKSP framework to learn inter-species complementarity and alleviating data limitations via increasing data diversity. Considering the training conflict arising from discrepancies in data distributions among species, we devise a Species-specific Batch Normalization (SBN) module, that involves multiple BN layers to separately fit the distributions of different species. To validate CKSP's effectiveness, experiments are performed on three public datasets from horses, sheep, and cattle, respectively. The results show that our approach remarkably boosts the classification performance compared to the baseline method (one-for-one framework) solely trained on individual-species data, with increments of 6.04%, 2.06%, and 3.66% in accuracy, and 10.33%, 3.67%, and 7.90% in F1-score for the horse, sheep, and cattle datasets, respectively. This proves the promising capabilities of our method in leveraging multi-species data to augment classification performance.

In this paper, we address three Principal--Agent problems in a moral hazard context and show that they are connected. We start by studying the problem of Principal with multiple Agents in cooperation. The term cooperation is manifested here by the fact that the agents optimize their criteria through Pareto equilibria. We show that as the number of agents tends to infinity, the principal's value function converges to the value function of a McKean--Vlasov control problem. Using the solution to this McKean--Vlasov control problem, we derive a constructive method for obtaining approximately optimal contracts for the principal's problem with multiple agents in cooperation. In a second step, we show that the problem of Principal with multiple Agents turns out to also converge, when the number of agents goes to infinity, towards a new Principal--Agent problem which is the Principal--Agent problem with Mckean--Vlasov dynamics. This is a Principal--Agent problem where the agent--controlled production follows a Mckean-Vlasov dynamics and the contract can depend of the distribution of the production. The value function of the principal in this setting is equivalent to that of the same McKean--Vlasov control problem from the multi--agent scenario. Furthermore, we show that an optimal contract can be constructed from the solution to this McKean--Vlasov control problem. We conclude by discussing, in a simple example, the connection of these problems with the multitask Principal--Agent problem which is a situation when a principal delegates multiple tasks that can be correlated to a single agent.

Inspired by the discovery of a variety of correlated insulators in the moir\'e universe, controlled by interactions projected to a set of isolated bands with a narrow bandwidth, we examine here a partial sum-rule associated with the inverse frequency-weighted optical conductivity restricted to low-energies. Unlike standard sum-rules that extend out to $infinite$ frequencies, which include contributions from $all$ inter-band transitions, we focus here on transitions associated $only$ with the $projected$ degrees of freedom. We analyze the partial sum-rule in a non-perturbative but "solvable" limit for a variety of correlation-induced insulators. This includes (i) magic-angle twisted bilayer graphene at integer-filling with projected Coulomb interactions, starting from the chiral flat-band limit and including realistic perturbations, (ii) fractional fillings of Chern-bands which support generalized Laughlin-like states, starting from a Landau-level and including a periodic potential and magnetic-field, respectively, drawing connections to twisted MoTe$_2$, and (iii) integer filling in toy-models of non-topological flat-bands with a tunable quantum geometry in the presence of repulsive interactions. The partial sum-rule in all of these examples is implicitly constrained by the form of the band quantum geometry via the low-lying excitation spectrum, but is not related to it explicitly. For interacting Slater-determinant insulators, the partial sum-rule is related to a new quantity -- "many-body projected quantum geometry" -- obtained from the interaction-renormalized electronic bands. We also point out an intriguing connection between the partial sum-rule and the quantum Fisher information associated with the projected many-body position operator., Comment: Main text: 14 pages, 5 figures, Supplementary information: 5 pages

Large language models (LLMs) have empowered nodes within multi-agent networks with intelligence, showing growing applications in both academia and industry. However, how to prevent these networks from generating malicious information remains unexplored with previous research on single LLM's safety be challenging to transfer. In this paper, we focus on the safety of multi-agent networks from a topological perspective, investigating which topological properties contribute to safer networks. To this end, we propose a general framework, NetSafe along with an iterative RelCom interaction to unify existing diverse LLM-based agent frameworks, laying the foundation for generalized topological safety research. We identify several critical phenomena when multi-agent networks are exposed to attacks involving misinformation, bias, and harmful information, termed as Agent Hallucination and Aggregation Safety. Furthermore, we find that highly connected networks are more susceptible to the spread of adversarial attacks, with task performance in a Star Graph Topology decreasing by 29.7%. Besides, our proposed static metrics aligned more closely with real-world dynamic evaluations than traditional graph-theoretic metrics, indicating that networks with greater average distances from attackers exhibit enhanced safety. In conclusion, our work introduces a new topological perspective on the safety of LLM-based multi-agent networks and discovers several unreported phenomena, paving the way for future research to explore the safety of such networks.

As large language model (LLM) agents increasingly integrate into our infrastructure, their robust coordination and message synchronization become vital. The Byzantine Generals Problem (BGP) is a critical model for constructing resilient multi-agent systems (MAS) under adversarial attacks. It describes a scenario where malicious agents with unknown identities exist in the system-situations that, in our context, could result from LLM agents' hallucinations or external attacks. In BGP, the objective of the entire system is to reach a consensus on the action to be taken. Traditional BGP requires global consensus among all agents; however, in practical scenarios, global consensus is not always necessary and can even be inefficient. Therefore, there is a pressing need to explore a refined version of BGP that aligns with the local coordination patterns observed in MAS. We refer to this refined version as Imperfect BGP (IBGP) in our research, aiming to address this discrepancy. To tackle this issue, we propose a framework that leverages consensus protocols within general MAS settings, providing provable resilience against communication attacks and adaptability to changing environments, as validated by empirical results. Additionally, we present a case study in a sensor network environment to illustrate the practical application of our protocol.

Recent advances in 1-bit Large Language Models (LLMs), such as BitNet and BitNet b1.58, present a promising approach to enhancing the efficiency of LLMs in terms of speed and energy consumption. These developments also enable local LLM deployment across a broad range of devices. In this work, we introduce bitnet.cpp, a tailored software stack designed to unlock the full potential of 1-bit LLMs. Specifically, we develop a set of kernels to support fast and lossless inference of ternary BitNet b1.58 LLMs on CPUs. Extensive experiments demonstrate that bitnet.cpp achieves significant speedups, ranging from 2.37x to 6.17x on x86 CPUs and from 1.37x to 5.07x on ARM CPUs, across various model sizes. The code is available at https://github.com/microsoft/BitNet.

As a cornerstone of modern information access, search engines have become indispensable in everyday life. With the rapid advancements in AI and natural language processing (NLP) technologies, particularly large language models (LLMs), search engines have evolved to support more intuitive and intelligent interactions between users and systems. Conversational search, an emerging paradigm for next-generation search engines, leverages natural language dialogue to facilitate complex and precise information retrieval, thus attracting significant attention. Unlike traditional keyword-based search engines, conversational search systems enhance user experience by supporting intricate queries, maintaining context over multi-turn interactions, and providing robust information integration and processing capabilities. Key components such as query reformulation, search clarification, conversational retrieval, and response generation work in unison to enable these sophisticated interactions. In this survey, we explore the recent advancements and potential future directions in conversational search, examining the critical modules that constitute a conversational search system. We highlight the integration of LLMs in enhancing these systems and discuss the challenges and opportunities that lie ahead in this dynamic field. Additionally, we provide insights into real-world applications and robust evaluations of current conversational search systems, aiming to guide future research and development in conversational search., Comment: 35 pages, 8 figures, continue to update

We present OpenMU-Bench, a large-scale benchmark suite for addressing the data scarcity issue in training multimodal language models to understand music. To construct OpenMU-Bench, we leveraged existing datasets and bootstrapped new annotations. OpenMU-Bench also broadens the scope of music understanding by including lyrics understanding and music tool usage. Using OpenMU-Bench, we trained our music understanding model, OpenMU, with extensive ablations, demonstrating that OpenMU outperforms baseline models such as MU-Llama. Both OpenMU and OpenMU-Bench are open-sourced to facilitate future research in music understanding and to enhance creative music production efficiency., Comment: Resources: https://github.com/sony/openmu

We study uniquely decodable codes and list decodable codes in the high-noise regime, specifically codes that are uniquely decodable from $\frac{1-\varepsilon}{2}$ fraction of errors and list decodable from $1-\varepsilon$ fraction of errors. We present several improved explicit constructions that achieve near-optimal rates, as well as efficient or even linear-time decoding algorithms. Our contributions are as follows. 1. Explicit Near-Optimal Linear Time Uniquely Decodable Codes: We construct a family of explicit $\mathbb{F}_2$-linear codes with rate $\Omega(\varepsilon)$ and alphabet size $2^{\mathrm{poly} \log(1/\varepsilon)}$, that are capable of correcting $e$ errors and $s$ erasures whenever $2e + s < (1 - \varepsilon)n$ in linear-time. 2. Explicit Near-Optimal List Decodable Codes: We construct a family of explicit list decodable codes with rate $\Omega(\varepsilon)$ and alphabet size $2^{\mathrm{poly} \log(1/\varepsilon)}$, that are capable of list decoding from $1-\varepsilon$ fraction of errors with a list size $L = \exp\exp\exp(\log^{\ast}n)$ in polynomial time. 3. List Decodable Code with Near-Optimal List Size: We construct a family of explicit list decodable codes with an optimal list size of $O(1/\varepsilon)$, albeit with a suboptimal rate of $O(\varepsilon^2)$, capable of list decoding from $1-\varepsilon$ fraction of errors in polynomial time. Furthermore, we introduce a new combinatorial object called multi-set disperser, and use it to give a family of list decodable codes with near-optimal rate $\frac{\varepsilon}{\log^2(1/\varepsilon)}$ and list size $\frac{\log^2(1/\varepsilon)}{\varepsilon}$, that can be constructed in probabilistic polynomial time and decoded in deterministic polynomial time. We also introduce new decoding algorithms that may prove valuable for other graph-based codes., Comment: 28 pages. To appear in SODA 2025

Critique ability, a meta-cognitive capability of humans, presents significant challenges for LLMs to improve. Recent works primarily rely on supervised fine-tuning (SFT) using critiques generated by a single LLM like GPT-4. However, these model-generated critiques often exhibit flaws due to the inherent complexity of the critique. Consequently, fine-tuning LLMs on such flawed critiques typically limits the model's performance and propagates these flaws into the learned model. To overcome these challenges, this paper proposes a novel data generation pipeline, named MultiCritique, that improves the critique ability of LLMs by utilizing multi-agent feedback in both the SFT and reinforcement learning (RL) stages. First, our data generation pipeline aggregates high-quality critiques from multiple agents instead of a single model, with crucial information as input for simplifying the critique. Furthermore, our pipeline improves the preference accuracy of critique quality through multi-agent feedback, facilitating the effectiveness of RL in improving the critique ability of LLMs. Based on our proposed MultiCritique data generation pipeline, we construct the MultiCritiqueDataset for the SFT and RL fine-tuning stages. Extensive experimental results on two benchmarks demonstrate: 1) the superior quality of our constructed SFT dataset compared to existing critique datasets; 2) additional improvements to the critique ability of LLMs brought by the RL stage. Notably, our fine-tuned 7B model significantly surpasses other advanced 7B-13B open-source models, approaching the performance of advanced 70B LLMs and GPT-4. Codes, datasets and model weights will be publicly available.

Federated Learning(FL) is popular as a privacy-preserving machine learning paradigm for generating a single model on decentralized data. However, statistical heterogeneity poses a significant challenge for FL. As a subfield of FL, personalized FL (pFL) has attracted attention for its ability to achieve personalized models that perform well on non-independent and identically distributed (Non-IID) data. However, existing pFL methods are limited in terms of leveraging the global model's knowledge to enhance generalization while achieving personalization on local data. To address this, we proposed a new method personalized Federated learning with Adaptive Feature Aggregation and Knowledge Transfer (FedAFK), to train better feature extractors while balancing generalization and personalization for participating clients, which improves the performance of personalized models on Non-IID data. We conduct extensive experiments on three datasets in two widely-used heterogeneous settings and show the superior performance of our proposed method over thirteen state-of-the-art baselines.

Approximate Nearest Neighbor Search (ANNS) is now widely used in various applications, ranging from information retrieval, question answering, and recommendation, to search for similar high-dimensional vectors. As the amount of vector data grows continuously, it becomes important to support updates to vector index, the enabling technique that allows for efficient and accurate ANNS on vectors. Because of the curse of high dimensionality, it is often costly to identify the right neighbors of a single new vector, a necessary process for index update. To amortize update costs, existing systems maintain a secondary index to accumulate updates, which are merged by the main index by global rebuilding the entire index periodically. However, this approach has high fluctuations of search latency and accuracy, not even to mention that it requires substantial resources and is extremely time-consuming for rebuilds. We introduce SPFresh, a system that supports in-place vector updates. At the heart of SPFresh is LIRE, a lightweight incremental rebalancing protocol to split vector partitions and reassign vectors in the nearby partitions to adapt to data distribution shift. LIRE achieves low-overhead vector updates by only reassigning vectors at the boundary between partitions, where in a high-quality vector index the amount of such vectors are deemed small. With LIRE, SPFresh provides superior query latency and accuracy to solutions based on global rebuild, with only 1% of DRAM and less than 10% cores needed at the peak compared to the state-of-the-art, in a billion scale vector index with 1% of daily vector update rate., Comment: SOSP 23