Searcharxiv⌕ Search

arXiv subjects

Jun Xu

Publications and source records attributed to Jun Xu.

At least 253 records · Page 14Linked to original sources
arXiv2023

Entity-Aspect-Opinion-Sentiment Quadruple Extraction for Fine-grained Sentiment Analysis

Product reviews often contain a large number of implicit aspects and object-attribute co-existence cases. Unfortunately, many existing studies in Aspect-Based Sentiment Analysis (ABSA) have overlooked this issue, which can make it difficult to extract opinions comprehensively and fairly. In this paper, we propose a new task called Entity-Aspect-Opinion-Sentiment Quadruple Extraction (EASQE), which aims to hierarchically decompose aspect terms into entities and aspects to avoid information loss, non-exclusive annotations, and opinion misunderstandings in ABSA tasks. To facilitate research in this new task, we have constructed four datasets (Res14-EASQE, Res15-EASQE, Res16-EASQE, and Lap14-EASQE) based on the SemEval Restaurant and Laptop datasets. We have also proposed a novel two-stage sequence-tagging based Trigger-Opinion framework as the baseline for the EASQE task. Empirical evaluations show that our Trigger-Opinion framework can generate satisfactory EASQE results and can also be applied to other ABSA tasks, significantly outperforming state-of-the-art methods. We have made the four datasets and source code of Trigger-Opinion publicly available to facilitate further research in this area.

cs.CL↗
arXiv2023

Scale-Adaptive Feature Aggregation for Efficient Space-Time Video Super-Resolution

The Space-Time Video Super-Resolution (STVSR) task aims to enhance the visual quality of videos, by simultaneously performing video frame interpolation (VFI) and video super-resolution (VSR). However, facing the challenge of the additional temporal dimension and scale inconsistency, most existing STVSR methods are complex and inflexible in dynamically modeling different motion amplitudes. In this work, we find that choosing an appropriate processing scale achieves remarkable benefits in flow-based feature propagation. We propose a novel Scale-Adaptive Feature Aggregation (SAFA) network that adaptively selects sub-networks with different processing scales for individual samples. Experiments on four public STVSR benchmarks demonstrate that SAFA achieves state-of-the-art performance. Our SAFA network outperforms recent state-of-the-art methods such as TMNet and VideoINR by an average improvement of over 0.5dB on PSNR, while requiring less than half the number of parameters and only 1/3 computational costs.

cs.CV↗
arXiv2023

4K-Resolution Photo Exposure Correction at 125 FPS with ~8K Parameters

The illumination of improperly exposed photographs has been widely corrected using deep convolutional neural networks or Transformers. Despite with promising performance, these methods usually suffer from large parameter amounts and heavy computational FLOPs on high-resolution photographs. In this paper, we propose extremely light-weight (with only ~8K parameters) Multi-Scale Linear Transformation (MSLT) networks under the multi-layer perception architecture, which can process 4K-resolution sRGB images at 125 Frame-Per-Second (FPS) by a Titan RTX GPU. Specifically, the proposed MSLT networks first decompose an input image into high and low frequency layers by Laplacian pyramid techniques, and then sequentially correct different layers by pixel-adaptive linear transformation, which is implemented by efficient bilateral grid learning or 1x1 convolutions. Experiments on two benchmark datasets demonstrate the efficiency of our MSLTs against the state-of-the-arts on photo exposure correction. Extensive ablation studies validate the effectiveness of our contributions. The code is available at https://github.com/Zhou-Yijie/MSLTNet.

cs.CV↗
arXiv2023

Reinforcement Re-ranking with 2D Grid-based Recommendation Panels

Modern recommender systems usually present items as a streaming, one-dimensional ranking list. Recently there is a trend in e-commerce that the recommended items are organized grid-based panels with two dimensions where users can view the items in both vertical and horizontal directions. Presenting items in grid-based result panels poses new challenges to recommender systems because existing models are all designed to output sequential lists while the slots in a grid-based panel have no explicit order. Directly converting the item rankings into grids (e.g., pre-defining an order on the slots) overlooks the user-specific behavioral patterns on grid-based panels and inevitably hurts the user experiences. To address this issue, we propose a novel Markov decision process (MDP) to place the items in 2D grid-based result panels at the final re-ranking stage of the recommender systems. The model, referred to as Panel-MDP, takes an initial item ranking from the early stages as the input. Then, it defines \emph{the MDP discrete time steps as the ranks in the initial ranking list, and the actions as the prediction of the user-item preference and the selection of the slots}. At each time step, Panel-MDP sequentially executes two sub-actions: first deciding whether the current item in the initial ranking list is preferred by the user; then selecting a slot for placing the item if preferred, or skipping the item otherwise. The process is continued until all of the panel slots are filled. The reinforcement learning algorithm of PPO is employed to implement and learn the parameters in the Panel-MDP. Simulation and experiments on a dataset collected from a widely-used e-commerce app demonstrated the superiority of Panel-MDP in terms of recommending 2D grid-based result panels.

cs.IR↗
arXiv2023

Reconciling results of 2019 and 2020 stellar occultations on Pluto's atmosphere. New constraints from both the 5 September 2019 event and consistency analysis

A stellar occultation by Pluto on 5 September 2019 yielded positive detections at two separate stations. Using an approach consistent with comparable studies, we derived a surface pressure of $11.478 \pm 0.55~\mathrm{μbar}$ for Pluto's atmosphere from the observations of this event. In addition, to avoid potential method inconsistancies highlighted by Sicardy et al. when comparing with historical pressure measurements, we reanalyzed the data by 15 August 2018 and 17 July 2019 events, respectively. All the new measurements provide a bridge between the two different perspectives on the pressure variation since 2015: a rapid pressure drop from previous studies of the 15 August 2018 and 17 July 2019 events and a plateau phase from that of the 6 June 2020 event. The pressure measurement from the 5 September 2019 event aligns with those from 2016, 2018, and 2020, supporting the latter perspective. While the measurements from the 4 June 2011 and 17 July 2019 events suggest probable V-shaped pressure variations unaccounted for by the volatile transport model (VTM) from Meza et al., the VTM remains applicable on average. And, the validity of the V-shaped variations is debatable due to the stellar faintness of the 4 June 2011 event and the grazing single-chord geometry of the 17 July 2019 event. To reveal and understand all significant pressure variations of Pluto's atmosphere, it is essential to provide constraints on both short-term and long-term evolutions of the interacting atmosphere and surface by continuous pressure monitoring through occultation observations, whenever possible, complemented by frequent spectroscopy and photometry of the surface.

astro-ph.EP↗
arXiv2023

The FAST Galactic Plane Pulsar Snapshot Survey: III. Timing results of 30 FAST-GPPS discovered pulsars

Timing observations are crucial for determining the basic parameters of newly discovered pulsars. Using the Five-hundred-meter Aperture Spherical radio Telescope (FAST) with the L-band 19-beam receiver covering the frequency range of 1.0--1.5 GHz, the FAST Galactic Plane Pulsar Snapshot (GPPS) Survey has discovered more than 600 faint pulsars with flux densities of only a few or a few tens of $μ$Jy at 1.25 GHz. To obtain accurate position, spin parameters and dispersion measure of a pulsar, and to calculate derived parameters such as the characteristic age and surface magnetic field, we collect available FAST pulsar data obtained either through targeted follow-up observations or through coincidental survey observations with one of the 19 beams of the receiver. From these data we obtain time of arrival (TOA) measurements for 30 newly discovered pulsars as well as for 13 known pulsars. We demonstrate that the TOA measurements acquired by the FAST from any beams of the receiver in any observation mode (e.g. the tracking mode or the snapshot mode) can be combined to get timing solutions. We update the ephemerides of 13 previously known pulsars and obtain the first phase-coherent timing results for 30 isolated pulsars discovered in the FAST GPPS Survey. Notably, PSR J1904+0853 is an isolated millisecond pulsar, PSR J1906+0757 is a disrupted recycled pulsar, and PSR J1856+0211 has a long period of 9.89 s that can constrain pulsar death lines. Based on these timing solutions, all available FAST data have been added together to obtain the best pulse profiles for these pulsars.

astro-ph.HE↗
arXiv2023

Reward Imputation with Sketching for Contextual Batched Bandits

Contextual batched bandit (CBB) is a setting where a batch of rewards is observed from the environment at the end of each episode, but the rewards of the non-executed actions are unobserved, resulting in partial-information feedback. Existing approaches for CBB often ignore the rewards of the non-executed actions, leading to underutilization of feedback information. In this paper, we propose an efficient approach called Sketched Policy Updating with Imputed Rewards (SPUIR) that completes the unobserved rewards using sketching, which approximates the full-information feedbacks. We formulate reward imputation as an imputation regularized ridge regression problem that captures the feedback mechanisms of both executed and non-executed actions. To reduce time complexity, we solve the regression problem using randomized sketching. We prove that our approach achieves an instantaneous regret with controllable bias and smaller variance than approaches without reward imputation. Furthermore, our approach enjoys a sublinear regret bound against the optimal policy. We also present two extensions, a rate-scheduled version and a version for nonlinear rewards, making our approach more practical. Experimental results show that SPUIR outperforms state-of-the-art baselines on synthetic, public benchmark, and real-world datasets.

cs.LG↗
arXiv2023

The FAST Galactic Plane Pulsar Snapshot Survey: II. Discovery of 76 Galactic rotating radio transients and their enigma

We are carrying out the GPPS survey by using the FAST, the most sensitive systematic pulsar survey in the Galactic plane. In addition to about 500 pulsars already discovered through normal periodical search, we report here the discovery of 76 new transient radio sources with sporadic strong pulses, detected by using the newly developed module for a sensitive single pulse search. Their small DM values suggest that they all are the Galactic RRATs. More radio pulses have been detected from 26 transient radio sources but no periods can be found due to a limited small number of pulses from all FAST observations. The following-up observations show that 16 transient sources are newly identified as being the prototypes of RRATs with a period already determined from more detected sporadic pulses, 10 sources are extremely nulling pulsars, and 24 sources are weak pulsars with sparse strong pulses. On the other hand, 48 previously known RRATs have been detected by the FAST. Except for 1 RRAT with four pulses detected in a session of five minute observation and 4 RRATs with only one pulse detected in a session, sensitive FAST observations reveal that 43 RRATs are just generally weak pulsars with sporadic strong pulses or simply very nulling pulsars, so that the previously known RRATs always have an extreme emission state together with a normal hardly detectable weak emission state. This is echoed by the two normal pulsars J1938+2213 and J1946+1449 with occasional brightening pulses. Though strong pulses of RRATs are very outstanding in the energy distribution, their polarization angle variations follow the polarization angle curve of the averaged normal pulse profile, suggesting that the predominant sparse pulses of RRATs are emitted in the same region with the same geometry as normal weak pulsars.

astro-ph.HE↗
arXiv2023

SYSPART: Automated Temporal System Call Filtering for Binaries

Restricting the system calls available to applications reduces the attack surface of the kernel and limits the functionality available to compromised applications. Recent approaches automatically identify the system calls required by programs to block unneeded ones. For servers, they even consider different phases of execution to tighten restrictions after initialization completes. However, they require access to the source code for applications and libraries, depend on users identifying when the server transitions from initialization to serving clients, or do not account for dynamically-loaded libraries. This paper introduces SYSPART, an automatic system-call filtering system designed for binary-only server programs that addresses the above limitations. Using a novel algorithm that combines static and dynamic analysis, SYSPART identifies the serving phases of all working threads of a server. Static analysis is used to compute the system calls required during the various serving phases in a sound manner, and dynamic observations are only used to complement static resolution of dynamically-loaded libraries when necessary. We evaluated SYSPART using six popular servers on x86-64 Linux to demonstrate its effectiveness in automatically identifying serving phases, generating accurate system-call filters, and mitigating attacks. Our results show that SYSPART outperforms prior binary-only approaches and performs comparably to source-code approaches.

cs.CR↗
arXiv2023

Characterization and Modeling of Silicon-on-Insulator Lateral Bipolar Junction Transistors at Liquid Helium Temperature

Conventional silicon bipolars are not suitable for low-temperature operation due to the deterioration of current gain ($β$). In this paper, we characterize lateral bipolar junction transistors (LBJTs) fabricated on silicon-on-insulator (SOI) wafers down to liquid helium temperature (4 K). The positive SOI substrate bias could greatly increase the collector current and have a negligible effect on the base current, which significantly alleviates $β$ degradation at low temperatures. We present a physical-based compact LBJT model for 4 K simulation, in which the collector current ($\textit{I}_\textbf{C}$) consists of the tunneling current and the additional current component near the buried oxide (BOX)/silicon interface caused by the substrate modulation effect. This model is able to fit the Gummel characteristics of LBJTs very well and has promising applications in amplifier circuits simulation for silicon-based qubits signals.

physics.app-ph↗
arXiv2023

Nanoscale Cathodoluminescence Spectroscopy Probing the Nitride Quantum Wells in an Electron Microcope

To gain a deeper understanding of the luminescence of multiquantum wells and the factors affecting it on a microscopic level, cathodoluminescence combined with scanning transmission electron microscopy and spectroscopy was used to reveal the luminescence of In0.15Ga0.85N five-period multiquantum wells. The composition-wave-energy relationship was established in combination with energy-dispersive X-ray spectroscopy , and the bandgaps of In0.15Ga0.85N and GaN in multiple quantum wells were extracted by electron energy loss spectroscopy to understand the features of cathodoluminescence luminescence spectra. The luminescence differences between different periods of multiquantum wells and the effects on the luminescence of multiple quantum wells owing to defects such as composition fluctuation and dislocations were revealed. Our study establishing the direct correspondence between the atomic structure of InxGa1-xN multiquantum wells and photoelectric properties, provides useful information for nitride applications.

physics.app-ph↗
arXiv2023

FCNet: A Convolutional Neural Network for Arbitrary-Length Exposure Estimation

The photographs captured by digital cameras usually suffer from over or under exposure problems. For image exposure enhancement, the tasks of Single-Exposure Correction (SEC) and Multi-Exposure Fusion (MEF) are widely studied in the image processing community. However, current SEC or MEF methods are developed under different motivations and thus ignore the internal correlation between SEC and MEF, making it difficult to process arbitrary-length sequences with improper exposures. Besides, the MEF methods usually fail at estimating the exposure of a sequence containing only under-exposed or over-exposed images. To alleviate these problems, in this paper, we develop a novel Fusion-Correction Network (FCNet) to tackle an arbitrary-length (including one) image sequence with improper exposures. This is achieved by fusing and correcting an image sequence by Laplacian Pyramid (LP) image decomposition. In each LP level, the low-frequency base component of the input image sequence is fed into a Fusion block and a Correction block sequentially for consecutive exposure estimation, implemented by alternative exposure fusion and correction. The exposure-corrected image in current LP level is upsampled and fused with the high-frequency detail components of the input image sequence in the next LP level, to output the base component for the Fusion and Correction blocks in next LP level. Experiments on the benchmark dataset demonstrate that our FCNet is effective on arbitrary-length exposure estimation, including both SEC and MEF. The code is publicly released at https://github.com/NKUJinLiang/FCNet.

eess.IV↗
arXiv2023

Joint Super-Resolution and Inverse Tone-Mapping: A Feature Decomposition Aggregation Network and A New Benchmark

Joint Super-Resolution and Inverse Tone-Mapping (joint SR-ITM) aims to increase the resolution and dynamic range of low-resolution and standard dynamic range images. Recent networks mainly resort to image decomposition techniques with complex multi-branch architectures. However, the fixed decomposition techniques would largely restricts their power on versatile images. To exploit the potential power of decomposition mechanism, in this paper, we generalize it from the image domain to the broader feature domain. To this end, we propose a lightweight Feature Decomposition Aggregation Network (FDAN). In particular, we design a Feature Decomposition Block (FDB) to achieve learnable separation of detail and base feature maps, and develop a Hierarchical Feature Decomposition Group by cascading FDBs for powerful multi-level feature decomposition. Moreover, to better evaluate the comparison methods, we collect a large-scale dataset for joint SR-ITM, i.e., SRITM-4K, which provides versatile scenarios for robust model training and evaluation. Experimental results on two benchmark datasets demonstrate that our FDAN is efficient and outperforms state-of-the-art methods on joint SR-ITM. The code of our FDAN and the SRITM-4K dataset are available at https://github.com/CS-GangXu/FDAN.

cs.CV↗
arXiv2023

Uncovering ChatGPT's Capabilities in Recommender Systems

The debut of ChatGPT has recently attracted the attention of the natural language processing (NLP) community and beyond. Existing studies have demonstrated that ChatGPT shows significant improvement in a range of downstream NLP tasks, but the capabilities and limitations of ChatGPT in terms of recommendations remain unclear. In this study, we aim to conduct an empirical analysis of ChatGPT's recommendation ability from an Information Retrieval (IR) perspective, including point-wise, pair-wise, and list-wise ranking. To achieve this goal, we re-formulate the above three recommendation policies into a domain-specific prompt format. Through extensive experiments on four datasets from different domains, we demonstrate that ChatGPT outperforms other large language models across all three ranking policies. Based on the analysis of unit cost improvements, we identify that ChatGPT with list-wise ranking achieves the best trade-off between cost and performance compared to point-wise and pair-wise ranking. Moreover, ChatGPT shows the potential for mitigating the cold start problem and explainable recommendation. To facilitate further explorations in this area, the full code and detailed original results are open-sourced at https://github.com/rainym00d/LLM4RS.

cs.IR↗
arXiv2023

Uncovering User Interest from Biased and Noised Watch Time in Video Recommendation

In the video recommendation, watch time is commonly adopted as an indicator of user interest. However, watch time is not only influenced by the matching of users' interests but also by other factors, such as duration bias and noisy watching. Duration bias refers to the tendency for users to spend more time on videos with longer durations, regardless of their actual interest level. Noisy watching, on the other hand, describes users taking time to determine whether they like a video or not, which can result in users spending time watching videos they do not like. Consequently, the existence of duration bias and noisy watching make watch time an inadequate label for indicating user interest. Furthermore, current methods primarily address duration bias and ignore the impact of noisy watching, which may limit their effectiveness in uncovering user interest from watch time. In this study, we first analyze the generation mechanism of users' watch time from a unified causal viewpoint. Specifically, we considered the watch time as a mixture of the user's actual interest level, the duration-biased watch time, and the noisy watch time. To mitigate both the duration bias and noisy watching, we propose Debiased and Denoised watch time Correction (D$^2$Co), which can be divided into two steps: First, we employ a duration-wise Gaussian Mixture Model plus frequency-weighted moving average for estimating the bias and noise terms; then we utilize a sensitivity-controlled correction function to separate the user interest from the watch time, which is robust to the estimation error of bias and noise terms. The experiments on two public video recommendation datasets and online A/B testing indicate the effectiveness of the proposed method.

cs.IR↗
arXiv2023

HyperBandit: Contextual Bandit with Hypernewtork for Time-Varying User Preferences in Streaming Recommendation

In real-world streaming recommender systems, user preferences often dynamically change over time (e.g., a user may have different preferences during weekdays and weekends). Existing bandit-based streaming recommendation models only consider time as a timestamp, without explicitly modeling the relationship between time variables and time-varying user preferences. This leads to recommendation models that cannot quickly adapt to dynamic scenarios. To address this issue, we propose a contextual bandit approach using hypernetwork, called HyperBandit, which takes time features as input and dynamically adjusts the recommendation model for time-varying user preferences. Specifically, HyperBandit maintains a neural network capable of generating the parameters for estimating time-varying rewards, taking into account the correlation between time features and user preferences. Using the estimated time-varying rewards, a bandit policy is employed to make online recommendations by learning the latent item contexts. To meet the real-time requirements in streaming recommendation scenarios, we have verified the existence of a low-rank structure in the parameter matrix and utilize low-rank factorization for efficient training. Theoretically, we demonstrate a sublinear regret upper bound against the best policy. Extensive experiments on real-world datasets show that the proposed HyperBandit consistently outperforms the state-of-the-art baselines in terms of accumulated rewards.

cs.IR↗
arXiv2023

KuaiSAR: A Unified Search And Recommendation Dataset

The confluence of Search and Recommendation (S&R) services is vital to online services, including e-commerce and video platforms. The integration of S&R modeling is a highly intuitive approach adopted by industry practitioners. However, there is a noticeable lack of research conducted in this area within academia, primarily due to the absence of publicly available datasets. Consequently, a substantial gap has emerged between academia and industry regarding research endeavors in joint optimization using user behavior data from both S&R services. To bridge this gap, we introduce the first large-scale, real-world dataset KuaiSAR of integrated Search And Recommendation behaviors collected from Kuaishou, a leading short-video app in China with over 350 million daily active users. Previous research in this field has predominantly employed publicly available semi-synthetic datasets and simulated, with artificially fabricated search behaviors. Distinct from previous datasets, KuaiSAR contains genuine user behaviors, including the occurrence of each interaction within either search or recommendation service, and the users' transitions between the two services. This work aids in joint modeling of S&R, and utilizing search data for recommender systems (and recommendation data for search engines). Furthermore, due to the various feedback labels associated with user-video interactions, KuaiSAR also supports a broad range of tasks, including intent recommendation, multi-task learning, and modeling of long sequential multi-behavioral patterns. We believe this dataset will serve as a catalyst for innovative research and bridge the gap between academia and industry in understanding the S&R services in practical, real-world applications.

cs.IR↗
arXiv2023

Information Retrieval Meets Large Language Models: A Strategic Report from Chinese IR Community

The research field of Information Retrieval (IR) has evolved significantly, expanding beyond traditional search to meet diverse user information needs. Recently, Large Language Models (LLMs) have demonstrated exceptional capabilities in text understanding, generation, and knowledge inference, opening up exciting avenues for IR research. LLMs not only facilitate generative retrieval but also offer improved solutions for user understanding, model evaluation, and user-system interactions. More importantly, the synergistic relationship among IR models, LLMs, and humans forms a new technical paradigm that is more powerful for information seeking. IR models provide real-time and relevant information, LLMs contribute internal knowledge, and humans play a central role of demanders and evaluators to the reliability of information services. Nevertheless, significant challenges exist, including computational costs, credibility concerns, domain-specific limitations, and ethical considerations. To thoroughly discuss the transformative impact of LLMs on IR research, the Chinese IR community conducted a strategic workshop in April 2023, yielding valuable insights. This paper provides a summary of the workshop's outcomes, including the rethinking of IR's core values, the mutual enhancement of LLMs and IR, the proposal of a novel IR technical paradigm, and open challenges.

cs.IR↗