Search

The ever-increasing size of language models curtails their widespread availability to the community, thereby galvanizing many companies into offering access to large language models through APIs. One particular type, suitable for dense retrieval, is a semantic embedding service that builds vector representations of input text. With a growing number of publicly available APIs, our goal in this paper is to analyze existing offerings in realistic retrieval scenarios, to assist practitioners and researchers in finding suitable services according to their needs. Specifically, we investigate the capabilities of existing semantic embedding APIs on domain generalization and multilingual retrieval. For this purpose, we evaluate these services on two standard benchmarks, BEIR and MIRACL. We find that re-ranking BM25 results using the APIs is a budget-friendly approach and is most effective in English, in contrast to the standard practice of employing them as first-stage retrievers. For non-English retrieval, re-ranking still improves the results, but a hybrid model with BM25 works best, albeit at a higher cost. We hope our work lays the groundwork for evaluating semantic embedding APIs that are critical in search and more broadly, for information access., Comment: ACL 2023 Industry Track

Supervised ranking methods based on bi-encoder or cross-encoder architectures have shown success in multi-stage text ranking tasks, but they require large amounts of relevance judgments as training data. In this work, we propose Listwise Reranker with a Large Language Model (LRL), which achieves strong reranking effectiveness without using any task-specific training data. Different from the existing pointwise ranking methods, where documents are scored independently and ranked according to the scores, LRL directly generates a reordered list of document identifiers given the candidate documents. Experiments on three TREC web search datasets demonstrate that LRL not only outperforms zero-shot pointwise methods when reranking first-stage retrieval results, but can also act as a final-stage reranker to improve the top-ranked results of a pointwise method for improved efficiency. Additionally, we apply our approach to subsets of MIRACL, a recent multilingual retrieval dataset, with results showing its potential to generalize across different languages.

Noise has always been nonnegligible trouble in object detection by creating confusion in model reasoning, thereby reducing the informativeness of the data. It can lead to inaccurate recognition due to the shift in the observed pattern, that requires a robust generalization of the models. To implement a general vision model, we need to develop deep learning models that can adaptively select valid information from multi-modal data. This is mainly based on two reasons. Multi-modal learning can break through the inherent defects of single-modal data, and adaptive information selection can reduce chaos in multi-modal data. To tackle this problem, we propose a universal uncertainty-aware multi-modal fusion model. It adopts a multi-pipeline loosely coupled architecture to combine the features and results from point clouds and images. To quantify the correlation in multi-modal information, we model the uncertainty, as the inverse of data information, in different modalities and embed it in the bounding box generation. In this way, our model reduces the randomness in fusion and generates reliable output. Moreover, we conducted a completed investigation on the KITTI 2D object detection dataset and its derived dirty data. Our fusion model is proven to resist severe noise interference like Gaussian, motion blur, and frost, with only slight degradation. The experiment results demonstrate the benefits of our adaptive fusion. Our analysis on the robustness of multi-modal fusion will provide further insights for future research.

Despite the prevalence of GPS services, they still suffer from intermittent positioning with poor accuracy in partially shadowed regions like urban canyons, flyover shadows, and factories' indoor areas. Existing wisdom relies on hardware modifications of GPS receivers or power-hungry infrastructures requiring continuous plug-in power supply which is hard to provide in outdoor regions and some factories. This paper fills the gap with GPSMirror, the first GPS-strengthening system that works for unmodified smartphones with the assistance of newly-designed GPS backscatter tags. The key enabling techniques in GPSMirror include: (i) a careful hardware design with microwatt-level power consumption that pushes the limit of backscatter sensitivity to re-radiate extremely weak GPS signals with enough coverage approaching the regulation limit; and (ii) a novel GPS positioning algorithm achieving meter-level accuracy in shadowed regions as well as expanding locatable regions under inadequate satellites where conventional algorithms fail. We build a prototype of the GPSMirror tags and conduct comprehensive experiments to evaluate them. Our results show that a GPSMirror tag can provide coverage up to 27.7 m. GPSMirror achieves median positioning accuracy of 3.7 m indoors and 4.6 m in urban canyon environments, respectively., Comment: 13 pages, 26 figures, to appear in MobiCom 2023

Causal inference is indispensable in many fields of empirical research, such as marketing, economic policy, medicine and so on. The estimation of treatment effects is the most important in causal inference. In this paper, we propose a model averaging approach, combined with a partition and matching method to estimate the conditional average treatment effect under heteroskedastic error settings. In our methods, we use the partition and matching method to approximate the true treatment effects and choose the weights by minimizing a leave-one-out cross validation criterion, which is also known as jackknife method. We prove that the model averaging estimator with weights determined by our criterion has asymptotic optimality, which achieves the lowest possible squared error. When there exist correct models in the candidate model set, we have proved that the sum of weights of the correct models converge to 1 as the sample size increases but with a finite number of baseline covariates. A Monte Carlo simulation study shows that our method has good performance in finite sample cases. We apply this approach to a National Supported Work Demonstration data set.

In this paper, we introduce a new NLP task -- generating short factual articles with references for queries by mining supporting evidence from the Web. In this task, called WebBrain, the ultimate goal is to generate a fluent, informative, and factually-correct short article (e.g., a Wikipedia article) for a factual query unseen in Wikipedia. To enable experiments on WebBrain, we construct a large-scale dataset WebBrain-Raw by extracting English Wikipedia articles and their crawlable Wikipedia references. WebBrain-Raw is ten times larger than the previous biggest peer dataset, which can greatly benefit the research community. From WebBrain-Raw, we construct two task-specific datasets: WebBrain-R and WebBrain-G, which are used to train in-domain retriever and generator, respectively. Besides, we empirically analyze the performances of the current state-of-the-art NLP techniques on WebBrain and introduce a new framework ReGen, which enhances the generation factualness by improved evidence retrieval and task-specific pre-training for generation. Experiment results show that ReGen outperforms all baselines in both automatic and human evaluations., Comment: Codes in https://github.com/qhjqhj00/WebBrain

Black-box adversarial attacks have shown strong potential to subvert machine learning models. Existing black-box attacks craft adversarial examples by iteratively querying the target model and/or leveraging the transferability of a local surrogate model. Recently, such attacks can be effectively mitigated by state-of-the-art (SOTA) defenses, e.g., detection via the pattern of sequential queries, or injecting noise into the model. To our best knowledge, we take the first step to study a new paradigm of black-box attacks with provable guarantees -- certifiable black-box attacks that can guarantee the attack success probability (ASP) of adversarial examples before querying over the target model. This new black-box attack unveils significant vulnerabilities of machine learning models, compared to traditional empirical black-box attacks, e.g., breaking strong SOTA defenses with provable confidence, constructing a space of (infinite) adversarial examples with high ASP, and the ASP of the generated adversarial examples is theoretically guaranteed without verification/queries over the target model. Specifically, we establish a novel theoretical foundation for ensuring the ASP of the black-box attack with randomized adversarial examples (AEs). Then, we propose several novel techniques to craft the randomized AEs while reducing the perturbation size for better imperceptibility. Finally, we have comprehensively evaluated the certifiable black-box attacks on the CIFAR10/100, ImageNet, and LibriSpeech datasets, while benchmarking with 16 SOTA empirical black-box attacks, against various SOTA defenses in the domains of computer vision and speech recognition. Both theoretical and experimental results have validated the significance of the proposed attack.

The concept of aerial-aquatic robots has emerged as an innovative solution that can operate both in the air and underwater. Previous research on the design of such robots has been mainly focused on mature technologies such as fixed-wing and multi-rotor aircraft. Flying fish, a unique aerial-aquatic animal that can both swim in water and glide over the sea surface, has not been fully explored as a bionic robot model, especially regarding its motion patterns with the collapsible pectoral fins. To verify the contribution of the collapsible wings to the flying fish motion pattern, we have designed a novel bio-robot with collapsible wings inspired by the flying fish. The bionic prototype has been successfully designed and fabricated, incorporating collapsible wings with soft hydraulic actuators, an innovative application of soft actuators to a micro aquatic-aerial robot. We have analyzed and built a precise model of dynamics for control, and tested both the soft hydraulic actuators and detailed aerodynamic coefficients. To further verify the feasibility of collapsible wings, we conducted simulations in different situations such as discharge angles, the area of collapsible wings, and the advantages of using ground effect. The results confirm the control of the collapsible wings and demonstrate the unique multi-modal motion pattern between water and air. Overall, our research represents the study of the collapsible wings in aquatic-aerial robots and significant contributes to the development of aquatic-aerial robots. The using of the collapsible wings must a contribution to the future aquatic-aerial robot.

The advent of multilingual language models has generated a resurgence of interest in cross-lingual information retrieval (CLIR), which is the task of searching documents in one language with queries from another. However, the rapid pace of progress has led to a confusing panoply of methods and reproducibility has lagged behind the state of the art. In this context, our work makes two important contributions: First, we provide a conceptual framework for organizing different approaches to cross-lingual retrieval using multi-stage architectures for mono-lingual retrieval as a scaffold. Second, we implement simple yet effective reproducible baselines in the Anserini and Pyserini IR toolkits for test collections from the TREC 2022 NeuCLIR Track, in Persian, Russian, and Chinese. Our efforts are built on a collaboration of the two teams that submitted the most effective runs to the TREC evaluation. These contributions provide a firm foundation for future advances.

In this paper, we propose a novel representation for grasping using contacts between multi-finger robotic hands and objects to be manipulated. This representation significantly reduces the prediction dimensions and accelerates the learning process. We present an effective end-to-end network, CMG-Net, for grasping unknown objects in a cluttered environment by efficiently predicting multi-finger grasp poses and hand configurations from a single-shot point cloud. Moreover, we create a synthetic grasp dataset that consists of five thousand cluttered scenes, 80 object categories, and 20 million annotations. We perform a comprehensive empirical study and demonstrate the effectiveness of our grasping representation and CMG-Net. Our work significantly outperforms the state-of-the-art for three-finger robotic hands. We also demonstrate that the model trained using synthetic data performs very well for real robots., Comment: The first two authors are with equal contributions. Paper accepted by ICRA 2023

Event-based cameras are bio-inspired sensors that capture brightness change of every pixel in an asynchronous manner. Compared with frame-based sensors, event cameras have microsecond-level latency and high dynamic range, hence showing great potential for object detection under high-speed motion and poor illumination conditions. Due to sparsity and asynchronism nature with event streams, most of existing approaches resort to hand-crafted methods to convert event data into 2D grid representation. However, they are sub-optimal in aggregating information from event stream for object detection. In this work, we propose to learn an event representation optimized for event-based object detection. Specifically, event streams are divided into grids in the x-y-t coordinates for both positive and negative polarity, producing a set of pillars as 3D tensor representation. To fully exploit information with event streams to detect objects, a dual-memory aggregation network (DMANet) is proposed to leverage both long and short memory along event streams to aggregate effective information for object detection. Long memory is encoded in the hidden state of adaptive convLSTMs while short memory is modeled by computing spatial-temporal correlation between event pillars at neighboring time intervals. Extensive experiments on the recently released event-based automotive detection dataset demonstrate the effectiveness of the proposed method.

While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric bird's eye view detection methods have inferior performances on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight, to address this issue. In essence, instead of predicting the pixel-wise depth, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. The code is available at {\url{https://github.com/ADLab-AutoDrive/BEVHeight}}., Comment: Accepted by CVPR 2023

Imitation learning aims to mimic the behavior of experts without explicit reward signals. Passive imitation learning methods which use static expert datasets typically suffer from compounding error, low sample efficiency, and high hyper-parameter sensitivity. In contrast, active imitation learning methods solicit expert interventions to address the limitations. However, recent active imitation learning methods are designed based on human intuitions or empirical experience without theoretical guarantee. In this paper, we propose a novel active imitation learning framework based on a teacher-student interaction model, in which the teacher's goal is to identify the best teaching behavior and actively affect the student's learning process. By solving the optimization objective of this framework, we propose a practical implementation, naming it AdapMen. Theoretical analysis shows that AdapMen can improve the error bound and avoid compounding error under mild conditions. Experiments on the MetaDrive benchmark and Atari 2600 games validate our theoretical analysis and show that our method achieves near-expert performance with much less expert involvement and total sampling steps than previous methods. The code is available at https://github.com/liuxhym/AdapMen.

We present Spacerini, a tool that integrates the Pyserini toolkit for reproducible information retrieval research with Hugging Face to enable the seamless construction and deployment of interactive search engines. Spacerini makes state-of-the-art sparse and dense retrieval models more accessible to non-IR practitioners while minimizing deployment effort. This is useful for NLP researchers who want to better understand and validate their research by performing qualitative analyses of training corpora, for IR researchers who want to demonstrate new retrieval models integrated into the growing Pyserini ecosystem, and for third parties reproducing the work of other researchers. Spacerini is open source and includes utilities for loading, preprocessing, indexing, and deploying search engines locally and remotely. We demonstrate a portfolio of 13 search engines created with Spacerini for different use cases.

Recent progress in information retrieval finds that embedding query and document representation into multi-vector yields a robust bi-encoder retriever on out-of-distribution datasets. In this paper, we explore whether late interaction, the simplest form of multi-vector, is also helpful to neural rerankers that only use the [CLS] vector to compute the similarity score. Although intuitively, the attention mechanism of rerankers at the previous layers already gathers the token-level information, we find adding late interaction still brings an extra 5% improvement in average on out-of-distribution datasets, with little increase in latency and no degradation in in-domain effectiveness. Through extensive experiments and analysis, we show that the finding is consistent across different model sizes and first-stage retrievers of diverse natures and that the improvement is more prominent on longer queries.

The deployment of 3D detectors strikes one of the major challenges in real-world self-driving scenarios. Existing BEV-based (i.e., Bird Eye View) detectors favor sparse convolutions (known as SPConv) to speed up training and inference, which puts a hard barrier for deployment, especially for on-device applications. In this paper, to tackle the challenge of efficient 3D object detection from an industry perspective, we devise a deployment-friendly pillar-based 3D detector, termed FastPillars. First, we introduce a novel lightweight Max-and-Attention Pillar Encoding (MAPE) module specially for enhancing small 3D objects. Second, we propose a simple yet effective principle for designing a backbone in pillar-based 3D detection. We construct FastPillars based on these designs, achieving high performance and low latency without SPConv. Extensive experiments on two large-scale datasets demonstrate the effectiveness and efficiency of FastPillars for on-device 3D detection regarding both performance and speed. Specifically, FastPillars delivers state-of-the-art accuracy on Waymo Open Dataset with 1.8X speed up and 3.8 mAPH/L2 improvement over CenterPoint (SPConv-based). Our code is publicly available at: https://github.com/StiphyJay/FastPillars., Comment: Submitted to AAAI2024

With the increasing complexity of the traffic environment, the importance of safety perception in intelligent driving is growing. Conventional methods in the robust perception of intelligent driving focus on training models with anomalous data, letting the deep neural network decide how to tackle anomalies. However, these models cannot adapt smoothly to the diverse and complex real-world environment. This paper proposes a new type of metric known as Eloss and offers a novel training strategy to empower perception models from the aspect of anomaly detection. Eloss is designed based on an explanation of the perception model's information compression layers. Specifically, taking inspiration from the design of a communication system, the information transmission process of an information compression network has two expectations: the amount of information changes steadily, and the information entropy continues to decrease. Then Eloss can be obtained according to the above expectations, guiding the update of related network parameters and producing a sensitive metric to identify anomalies while maintaining the model performance. Our experiments demonstrate that Eloss can deviate from the standard value by a factor over 100 with anomalous data and produce distinctive values for similar but different types of anomalies, showing the effectiveness of the proposed method. Our code is available at: (code available after paper accepted).

Modelling human cognition can provide key insights into behavioral dynamics under changing conditions. This enables synthetic data generation and guides adaptive interventions for cognitive regulation. Challenges arise when environments are highly dynamic, obscuring stimulus-behavior relationships. We propose a cognitive agent integrating drift-diffusion with deep reinforcement learning to simulate granular stress effects on logical reasoning process. Leveraging a large dataset of 21,157 logical responses, we investigate performance impacts of dynamic stress. This prior knowledge informed model design and evaluation. Quantitatively, the framework improves cognition modelling by capturing both subject-specific and stimuli-specific behavioural differences. Qualitatively, it captures general trends in human logical reasoning under stress. Our approach is extensible to examining diverse environmental influences on cognition and behavior. Overall, this work demonstrates a powerful, data-driven methodology to simulate and understand the vagaries of human logical reasoning process in dynamic contexts., Comment: 17 pages, 7 figures

Conventional methods for extreme event estimation rely on well-chosen parametric models asymptotically justified from extreme value theory (EVT). These methods, while powerful and theoretically grounded, could however encounter a difficult bias-variance tradeoff that exacerbates especially when data size is too small, deteriorating the reliability of the tail estimation. In this paper, we study a framework based on the recently surging literature of distributionally robust optimization. This approach can be viewed as a nonparametric alternative to conventional EVT, by imposing general shape belief on the tail instead of parametric assumption and using worst-case optimization as a resolution to handle the nonparametric uncertainty. We explain how this approach bypasses the bias-variance tradeoff in EVT. On the other hand, we face a conservativeness-variance tradeoff which we describe how to tackle. We also demonstrate computational tools for the involved optimization problems and compare our performance with conventional EVT across a range of numerical examples.

Biological systems in nature have evolved for millions of years to adapt and survive the environment. Many features they developed can be inspirational and beneficial for solving technical problems in modern industries. This leads to a specific form of design-by-analogy called bio-inspired design (BID). Although BID as a design method has been proven beneficial, the gap between biology and engineering continuously hinders designers from effectively applying the method. Therefore, we explore the recent advance of artificial intelligence (AI) for a data-driven approach to bridge the gap. This paper proposes a generative design approach based on the generative pre-trained language model (PLM) to automatically retrieve and map biological analogy and generate BID in the form of natural language. The latest generative pre-trained transformer, namely GPT-3, is used as the base PLM. Three types of design concept generators are identified and fine-tuned from the PLM according to the looseness of the problem space representation. Machine evaluators are also fine-tuned to assess the mapping relevancy between the domains within the generated BID concepts. The approach is evaluated and then employed in a real-world project of designing light-weighted flying cars during its conceptual design phase The results show our approach can generate BID concepts with good performance., Comment: Accepted by J. Mech. Des. arXiv admin note: substantial text overlap with arXiv:2204.09714

Multi-Hop Question Answering (MHQA) is a significant area in question answering, requiring multiple reasoning components, including document retrieval, supporting sentence prediction, and answer span extraction. In this work, we analyze the primary factors limiting the performance of multi-hop reasoning and introduce label smoothing into the MHQA task. This is aimed at enhancing the generalization capabilities of MHQA systems and mitigating overfitting of answer spans and reasoning paths in training set. We propose a novel label smoothing technique, F1 Smoothing, which incorporates uncertainty into the learning process and is specifically tailored for Machine Reading Comprehension (MRC) tasks. Inspired by the principles of curriculum learning, we introduce the Linear Decay Label Smoothing Algorithm (LDLA), which progressively reduces uncertainty throughout the training process. Experiment on the HotpotQA dataset demonstrates the effectiveness of our methods in enhancing performance and generalizability in multi-hop reasoning, achieving new state-of-the-art results on the leaderboard., Comment: 13 pages, 8 figures, accepted by CCL2023

Large crossed mixed effects models with imbalanced structures and missing data pose major computational challenges for standard Bayesian posterior sampling algorithms, as the computational complexity is usually superlinear in the number of observations. We propose two efficient subset-based stochastic gradient MCMC algorithms for such crossed mixed effects model, which facilitate scalable inference on both the variance components and regression coefficients. The first algorithm is developed for balanced design without missing observations, where we leverage the closed-form expression of precision matrix for the full data matrix. The second algorithm, which we call the pigeonhole stochastic gradient Langevin dynamics (PSGLD), is developed for both balanced and unbalanced designs with potentially a large proportion of missing observations. Our PSGLD algorithm imputes the latent crossed random effects by running short Markov chains and then samples the model parameters of variance components and regression coefficients at each MCMC iteration. We provide theoretical guarantee by showing the convergence of the output distribution from the proposed algorithms to the target non-log-concave posterior distribution. A variety of numerical experiments based on both synthetic and real data demonstrate that the proposed algorithms can significantly reduce the computational cost of the standard MCMC algorithms and better balance the approximation accuracy and computational efficiency.

The use of water-soluble sacrificial layer of Sr$_3$Al$_2$O$_6$ has tremendously boosted the research on freestanding functional oxide thin films, especially thanks to its ultimate capability to produce high-quality epitaxial perovskite thin films. However, the costly single-crystalline substrates, e.g. SrTiO$_3$, were generally discarded after obtaining the freestanding thin films. Here, we demonstrate that the SrTiO$_3$ substrates can be recycled to fabricate La$_{0.7}$Sr$_{0.3}$MnO$_3$ films with nearly identical structural and electrical properties. After attaining freestanding thin films, the residues on SrTiO$_3$ can be removed by 80 \degree C hot water soaking and rinsing treatments. Consequently, the surface of SrTiO$_3$ reverted to its original step-and-terrace structure.