The Retrieval-Augmented Language Model (RALM) has demonstrated remarkable performance on knowledge-intensive tasks by integrating external knowledge during inference, which mitigates the factual hallucinations inherited in large language models (LLMs). Despite these advancements, challenges persist in the implementation of RALMs, particularly in terms of reliability and traceability. Specifically, the irrelevant document retrieval may result in unhelpful responses or even deteriorate the performance of LLMs, while the lack of appropriate citations in outputs complicates efforts to verify the trustworthiness of the models. To this end, we propose a novel self-reasoning framework aimed at improving the reliability and traceability of RALMs, whose core idea is to leverage reasoning trajectories generated by the LLM itself. The framework involves constructing self-reasoning trajectories through three processes: a relevance-aware process, an evidence-aware selective process, and a trajectory analysis process. We evaluated our framework across four public datasets (two short-form QA datasets, one long-form QA dataset, and one fact verification dataset) to demonstrate its superiority. Our method can outperform existing state-of-the-art models and achieve performance comparable with GPT-4, using only 2,000 training samples.

With the proliferation of images in online content, language-guided image retrieval (LGIR) has emerged as a research hotspot over the past decade, encompassing a variety of subtasks with diverse input forms. While the development of large multimodal models (LMMs) has significantly facilitated these tasks, existing approaches often address them in isolation, requiring the construction of separate systems for each task. This not only increases system complexity and maintenance costs, but also exacerbates challenges stemming from language ambiguity and complex image content, making it difficult for retrieval systems to provide accurate and reliable results. To this end, we propose ImageScope, a training-free, three-stage framework that leverages collective reasoning to unify LGIR tasks. The key insight behind the unification lies in the compositional nature of language, which transforms diverse LGIR tasks into a generalized text-to-image retrieval process, along with the reasoning of LMMs serving as a universal verification to refine the results. To be specific, in the first stage, we improve the robustness of the framework by synthesizing search intents across varying levels of semantic granularity using chain-of-thought (CoT) reasoning. In the second and third stages, we then reflect on retrieval results by verifying predicate propositions locally, and performing pairwise evaluations globally. Experiments conducted on six LGIR datasets demonstrate that ImageScope outperforms competitive baselines. Comprehensive evaluations and ablation studies further confirm the effectiveness of our design.

Uncertainty in safety-critical cyber-physical systems can be modeled using a finite number of parameters or parameterized input signals. Given a system specification in Signal Temporal Logic (STL), we would like to verify that for all (infinite) values of the model parameters/input signals, the system satisfies its specification. Unfortunately, this problem is undecidable in general. Statistical model checking (SMC) offers a solution by providing guarantees on the correctness of CPS models by statistically reasoning on model simulations. We propose a new approach for statistical verification of CPS models for user-provided distribution on the model parameters. Our technique uses model simulations to learn surrogate models , and uses conformal inference to provide probabilistic guarantees on the satisfaction of a given STL property. Additionally, we can provide prediction intervals containing the quantitative satisfaction values of the given STL property for any user-specified confidence level. We compare this prediction interval with the interval we get using risk estimation procedures. We also propose a refinement procedure based on Gaussian Process (GP)-based surrogate models for obtaining fine-grained probabilistic guarantees over sub-regions in the parameter space. This in turn enables the CPS designer to choose assured validity domains in the parameter space for safety-critical applications. Finally, we demonstrate the efficacy of our technique on several CPS models.

Zone Namespace (ZNS) SSDs delegate data placement and garbage collection (GC) to the host, enabling applications on the host to perform efficient GC. The existing works on LSMT-based KV stores adopt ZenFS (a user-level file system) to manage ZNS SSDs. ZenFS assumes that SSTs within the same level have similar lifetimes and places SSTs with similar lifetimes into the same zone to minimize data migration in GC. However, we observe significant disparity in the lifetimes of SSTs within the same level, resulting in fragmented zones and huge data migration in GC. To reduce data migration in GC and improve performance, we present LifetimeKV, an LSMT-based KV store for ZNS SSDs. LifetimeKV introduces a range compaction algorithm to reduce short-lived SSTs and an overlap-ratio-lifetime victim SST selection algorithm to reduce long-lived SSTs, thereby reducing the lifetime disparity among SSTs within the same level. We evaluate LifetimeKV performance on a real ZNS SSD. The results demonstrate that LifetimeKV reduces data migration in GC by 63.23% and achieves a throughput improvement of 98.81% under write-intensive workload than state-of-the-art work.

Biological Knowledge Graphs (BKGs) can help to model complex biological systems in a structural way to support various tasks. Nevertheless, the incompleteness problem may limit the performance of existing BKGs, which still deserves new methods to reveal the missing relations. Though great efforts have been made to knowledge graph completion, existing methods are not easy to be adapted to the multimodal biological information such as molecular structures and textual descriptions. To this end, we propose a novel co-attention-based multimodal embedding framework, named CamE, for the multimodal BKG completion task. Specifically, we design a Triple Co-Attention (TCA) operator to capture and highlight the same semantic features among different modalities. Based on TCA, we further propose two components to handle multimodal fusion and multimodal entity-relation interaction, respectively. One is the multimodal TCA fusion module to achieve a multimodal joint representation for each entity in the BKG. It aims to project different modal information into a common space by capturing the same semantic features and overcoming the modality gap. The other is the relation-aware interactive TCA module to learn interactive representation by modelling the deep interaction between multimodal entities and relations. Extensive experiments on two real-world multimodal BKG datasets demonstrate that our method significantly outperforms several state-of-the-art baselines, including 10.3% and 16.2% improvement w.r.t MRR and Hits@1 metrics over its best competitors on public DRKG-MM dataset.

Yuan Xia, Zhenhui Shi, Jingbo Zhou, Jiayu Xu, Chao Lu, Yehui Yang, Lei Wang, Haifeng Huang, Xia Zhang, Junwei Liu. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. 2022.

Medical entity relation verification is a crucial step to build a practical and enterprise medical knowledge graph (MKG) because high-precision medical entity relation is a key requirement for many MKG-based applications. Existing relation verification approaches for general knowledge graphs are not designed for considering medical domain knowledge, although it is central to achieve high-quality entity relation verification for MKG. To this end, in this paper, we introduce a system for medical entity relation verification with large-scale machine reading comprehension. The proposed system is tailored to overcome the unique challenges of medical relation verification including high variants of medical terms, the high difficulty of evidence searching in complex medical documents, and the lack of evidence labels for supervision. To deal with the problem of variants of medical terms, we introduce a synonym-aware retrieve model to retrieve the potential evidence implicitly verifying the given claim. To better utilize the medical domain knowledge, a relation-aware evidence detector and a medical ontology-enhanced aggregator are developed to improve the performance of the relation verification module. Moreover, to overcome the challenge of providing high-quality evidence due to the lack of labels, we introduce an interactive collaborative-training method to iteratively improve the evidence accuracy. Finally, we conduct extensive experiments to demonstrate that the performance of our proposed system is superior to all comparable models. We also demonstrate that our system can significantly reduce the annotation time by medical experts in real-world verification tasks. It can help to improve the efficiency by nearly 300%. In particular, our system has been embedded into the Baidu Clinical Decision Support System.

Automatic diagnosis systems have attracted increasing attention in recent years. The reinforcement learning (RL) is an attractive technique for building an automatic diagnosis system due to its advantages for handling sequential decision making problem. However, the RL method still cannot achieve good enough prediction accuracy. In this paper, we propose a Generative Adversarial regularized Mutual information Policy gradient framework (GAMP) for automatic diagnosis which aims to make a diagnosis rapidly and accurately. We first propose a new policy gradient framework based on the Generative Adversarial Network (GAN) to optimize the RL model for automatic diagnosis. In our framework, we take the generator of GAN as a policy network, and also use the discriminator of GAN as a part of the reward function. This generative adversarial regularized policy gradient framework can try to avoid generating randomized trials of symptom inquires deviated from the common diagnosis paradigm. In addition, we add mutual information to enhance the reward function to encourage the model to select the most discriminative symptoms to make a diagnosis. Experiment evaluations on two public datasets show that our method beats the state-of-art methods, not only can achieve higher diagnosis accuracy, but also can use a smaller number of inquires to make diagnosis decision.

Self-driving cars, or Autonomous Vehicles (AVs), are increasingly becoming an integral part of our daily life. About 50 corporations are actively working on AVs, including large companies such as Google, Ford, and Intel. Some AVs are already operating on public roads, with at least one unfortunate fatality recently on record. As a result, understanding bugs in AVs is critical for ensuring their security, safety, robustness, and correctness. While previous studies have focused on a variety of domains (e.g., numerical software; machine learning; and error-handling, concurrency, and performance bugs) to investigate bug characteristics, AVs have not been studied in a similar manner. Recently, two software systems for AVs, Baidu Apollo and Autoware, have emerged as frontrunners in the open-source community and have been used by large companies and governments (e.g., Lincoln, Volvo, Ford, Intel, Hitachi, LG, and the US Department of Transportation). From these two leading AV software systems, this paper describes our investigation of 16,851 commits and 499 AV bugs and introduces our classification of those bugs into 13 root causes, 20 bug symptoms, and 18 categories of software components those bugs often affect. We identify 16 major findings from our study and draw broader lessons from them to guide the research community towards future directions in software bug detection, localization, and repair.

The complexity of biological data is a difficult challenge for data integration. Thus, this study develops a localized web service platform for the specific bioinformatics study needs of Southwest University of China. Our platform includes the following three innovation features, (1) Graphical Survival time analysis kit; (2) Codon Deviation Coefficient (CDC) computing kit and (3) Long non-coding RNA (lncRNA) analysis kit. These feature offer the convenience to the experimentalists for the biological data analysis.

In the above paper[1], there are two errors on page 7058. In addition, the support information has been updated to include an additional funding source. The complete funding statement should be:

A combined lidar system on the basis of conventional Rayleigh lidar has been extended by two rotational Raman (RR) channels for nocturnal atmospheric temperature measurements from 5 to 80 km over Wuhan, China (30.5°N, 114.5°E). An overlapping altitude range of about 10 km is obtained with the RR-technique temperatures reaching upward to 40 km and the Rayleigh-integration-technique temperatures above 30 km. Temperature values obtained by two different mechanisms match nicely in the overlapping area. By using a data-merge method, complete temperature profiles covering widely from 5 to 80 km are obtained for the observation of the thermal structure and perturbations from the troposphere up to the mesosphere. Based on the overlapping-region (30–40 km) data obtained from this combined RR–Rayleigh lidar system and Rayleigh-integration-technique temperatures initialized with model data at an upper height (90 km), we develop a self-calibration method for the determination of the system-dependent constants for RR temperature retrieval. Compared with the conventional radiosonde calibration method, the self-calibration obtained in the overlap region of both lidar temperature measurement techniques can be extrapolated to the lower temperatures in the tropopause region by using the simpler two-constant calibration function. With this new calibration method, the combined lidar system can perform independent and accurate atmospheric temperature measurements when a coincident (in time and space) radiosonde is not available, as it is often the case. This combined RR–Rayleigh lidar thus has the potential for long-term studies of atmospheric thermal structure and associate perturbations.

Multi-band OFDM Alliance (MBOA) Ultra-wide band (UWB) is a candidate of IEEE 802.15.3a, the standard for high speed Wireless Personal Area Networks (WPAN). The Media Access Control (MAC) protocol of MBOA consists of distributed reservation period (DRP) and Prioritized contention access (PCA), where DRP provides flexible distributed slot reservation for Quality of Service (QoS) support. This paper analyzes the delay performance of DRP channel access for MBOA UWB MAC. In DRP, the reserved periodical slot pattern in each super-frame for station can be arbitrary and not evenly distributed. To address the difference in delay performance introduced by reservation pattern, a bi-dimensional Markov chain model is proposed, where one dimension is for queue size distribution and the other is for allocated slots. The accuracy of the model is verified by simulations and insights has been given for slot reservation in MBOA.

Subcarrier and bit allocation schemes for SISO-OFDM systems in multiuser downlink scenario are well-documented. In this paper, we extend it to MIMO-OFDM systems and design a transmit scheme as a concatenation of dynamic subcarrier assignment, adaptive modulation and beamforming. With the goal of minimizing the overall transmit power required to meet the target BER, we develop an adaptive OFDM subcarrier allocation approach based on perfect and partial channel information. Simulation results show that our proposed algorithm outperforms the multiuser MIMO-OFDM with static FDMA technique considerably, and can lessen the performance loss caused by feedback delays. We have also evaluated the performance of the presented system under various number of transmit- and receive-antennas.