Facade parsing stands as a pivotal computer vision task with far-reaching applications in areas like architecture, urban planning, and energy efficiency. Despite the recent success of deep learning-based methods in yielding impressive results on certain open-source datasets, their viability for real-world applications remains uncertain. Real-world scenarios are considerably more intricate, demanding greater computational efficiency. Existing datasets often fall short in representing these settings, and previous methods frequently rely on extra models to enhance accuracy, which requires much computation cost. In this paper, we introduce Comprehensive Facade Parsing (CFP), a dataset meticulously designed to encompass the intricacies of real-world facade parsing tasks. Comprising a total of 602 high-resolution street-view images, this dataset captures a diverse array of challenging scenarios, including sloping angles and densely clustered buildings, with painstakingly curated annotations for each image. We introduce a new pipeline known as Revision-based Transformer Facade Parsing (RTFP). This marks the pioneering utilization of Vision Transformers (ViT) in facade parsing, and our experimental results definitively substantiate its merit. We also design Line Acquisition, Filtering, and Revision (LAFR), an efficient yet accurate revision algorithm that can improve the segment result solely from simple line detection using prior knowledge of the facade. In ECP 2011, RueMonge 2014, and our CFP, we evaluate the superiority of our method.
With the increasing size of pre-trained language models (PLMs), fine-tuning all the parameters in the model is not efficient, especially when there are a large number of downstream tasks, which incur significant training and storage costs. Many parameter-efficient fine-tuning (PEFT) approaches have been proposed, among which, Low-Rank Adaptation (LoRA) is a representative approach that injects trainable rank decomposition matrices into every target module. Yet LoRA ignores the importance of parameters in different modules. To address this problem, many works have been proposed to prune the parameters of LoRA. However, under limited training conditions, the upper bound of the rank of the pruned parameter matrix is still affected by the preset values. We, therefore, propose IncreLoRA, an incremental parameter allocation method that adaptively adds trainable parameters during training based on the importance scores of each module. This approach is different from the pruning method as it is not limited by the initial number of training parameters, and each parameter matrix has a higher rank upper bound for the same training overhead. We conduct extensive experiments on GLUE to demonstrate the effectiveness of IncreLoRA. The results show that our method owns higher parameter efficiency, especially when under the low-resource settings where our method significantly outperforms the baselines. Our code is publicly available.
The recent advances in NLP, have led to a new trend of applying LLMs to real-world scenarios. While the latest LLMs are astonishingly fluent when interacting with humans, they suffer from the misinformation problem by unintentionally generating factually false statements. This can lead to harmful consequences, especially when produced within sensitive contexts, such as healthcare. Yet few previous works have focused on evaluating misinformation in the long-form generation of LLMs, especially for knowledge-intensive topics. Moreover, although LLMs have been shown to perform well in different languages, misinformation evaluation has been mostly conducted in English. To this end, we present a benchmark, CARE-MI, for evaluating LLM misinformation in: 1) a sensitive topic, specifically the maternity and infant care domain; and 2) a language other than English, namely Chinese. Most importantly, we provide an innovative paradigm for building long-form generation evaluation benchmarks that can be transferred to other knowledge-intensive domains and low-resourced languages. Our proposed benchmark fills the gap between the extensive usage of LLMs and the lack of datasets for assessing the misinformation generated by these models. It contains 1,612 expert-checked questions, accompanied with human-selected references. Using our benchmark, we conduct extensive experiments and found that current Chinese LLMs are far from perfect in the topic of maternity and infant care. In an effort to minimize the reliance on human resources for performance evaluation, we offer a judgment model for automatically assessing the long-form output of LLMs using the benchmark questions. Moreover, we compare potential solutions for long-form generation evaluation and provide insights for building more robust and efficient automated metric.
The identification and discovery of drug-target Interaction (DTI) is an important step in the field of Drug research and development, which can help scientists discover new drugs and accelerate the development process. KnowledgeGraph and the related knowledge graph Embedding (KGE) model develop rapidly and show good performance in the field of drug discovery in recent years. In the task of drug target identification, the lack of authenticity and accuracy of the model will lead to the increase of misjudgment rate and the low efficiency of drug development. To solve the above problems, this study focused on the problem of drug target link prediction with knowledge mapping as the core technology, and adopted the confidence measurement method based on causal intervention to measure the triplet score, so as to improve the accuracy of drug target interaction prediction model. By comparing with the traditional Softmax and Sigmod confidence measurement methods on different KGE models, the results show that the confidence measurement method based on causal intervention can effectively improve the accuracy of DTI link prediction, especially for high-precision models. The predicted results are more conducive to guiding the design and development of followup experiments of drug development, so as to improve the efficiency of drug development.
The orthogonal time frequency space (OTFS) modulation as a promising signal representation attracts growingcinterest for integrated sensing and communication (ISAC), yet its merits over orthogonal frequency division multiplexing (OFDM) remain controversial. This paper devotes to a comprehensive comparison of OTFS and OFDM for sensing from the perspective of Cramer-Rao lower bounds (CRLB) analysis. To this end, we develop the cyclic prefix (CP)-Free and CP-added model for OFDM, while for OTFS, we consider the Zak transform based and the Two-Step conversion based models, respectively. Then we rephrase these four models into a unified matrix format to derive the CRLB of the delays and doppler shifts for multipath scenario. Numerical results demonstrate the superiority of OTFS modulation for sensing, and the effect of physical parameters for performance achievement.
Interpreting and explaining the behavior of deep neural networks is critical for many tasks. Explainable AI provides a way to address this challenge, mostly by providing per-pixel relevance to the decision. Yet, interpreting such explanations may require expert knowledge. Some recent attempts toward interpretability adopt a concept-based framework, giving a higher-level relationship between some concepts and model decisions. This paper proposes Bottleneck Concept Learner (BotCL), which represents an image solely by the presence/absence of concepts learned through training over the target task without explicit supervision over the concepts. It uses self-supervision and tailored regularizers so that learned concepts can be human-understandable. Using some image classification tasks as our testbed, we demonstrate BotCL's potential to rebuild neural networks for better interpretability. Code is available at https://github.com/wbw520/BotCL and a simple demo is available at https://botcl.liangzhili.com/.
Attaining the equilibrium state of a catalyst-adsorbate system is key to fundamentally assessing its effective properties, such as adsorption energy. Machine learning methods with finer supervision strategies have been applied to boost and guide the relaxation process of an atomic system and better predict its properties at the equilibrium state. In this paper, we present a novel graph neural network (GNN) supervision and prediction strategy DR-Label. The method enhances the supervision signal, reduces the multiplicity of solutions in edge representation, and encourages the model to provide node predictions that are graph structural variation robust. DR-Label first Deconstructs finer-grained equilibrium state information to the model by projecting the node-level supervision signal to each edge. Reversely, the model Reconstructs a more robust equilibrium state prediction by transforming edge-level predictions to node-level with a sphere-fitting algorithm. The DR-Label strategy was applied to three radically distinct models, each of which displayed consistent performance enhancements. Based on the DR-Label strategy, we further proposed DRFormer, which achieved a new state-of-the-art performance on the Open Catalyst 2020 (OC20) dataset and the Cu-based single-atom-alloyed CO adsorption (SAA) dataset. We expect that our work will highlight crucial steps for the development of a more accurate model in equilibrium state property prediction of a catalysis system.
Offline safe RL is of great practical relevance for deploying agents in real-world applications. However, acquiring constraint-satisfying policies from the fixed dataset is non-trivial for conventional approaches. Even worse, the learned constraints are stationary and may become invalid when the online safety requirement changes. In this paper, we present a novel offline safe RL approach referred to as SaFormer, which tackles the above issues via conditional sequence modeling. In contrast to existing sequence models, we propose cost-related tokens to restrict the action space and a posterior safety verification to enforce the constraint explicitly. Specifically, SaFormer performs a two-stage auto-regression conditioned by the maximum remaining cost to generate feasible candidates. It then filters out unsafe attempts and executes the optimal action with the highest expected return. Extensive experiments demonstrate the efficacy of SaFormer featuring (1) competitive returns with tightened constraint satisfaction; (2) adaptability to the in-range cost values of the offline data without retraining; (3) generalizability for constraints beyond the current dataset.
Hybrid digital/analog architecture and low-resolution analog-to-digital/digital-to-analog converters (ADCs /DACs) are two low-cost implementations for large-scale millimeter wave (mmWave) systems. In this paper, we investigate the problem of constant-envelope transmit beamforming for large-scale multiple-input multiple-output (MIMO) radar system, where the transmit array adopts a hybrid digital/analog architecture with a small number of RF chains and the receive array adopts a fully digital architecture with low-resolution ADCs. We derive the relative entropy between the probability density functions associated with the two test hypotheses under low-resolution ADCs. We formulate our optimization problem by maximizing the relative entropy, subject to the constant envelope and orthogonality constraints. To suboptimally solve the resultant problem, a two-stage framework is developed. In the first stage, we optimize the transmit power at the directions of the target and clutter. In the second stage, an efficient iterative algorithm based on majorization-minimization is presented to obtain the constant-envelope beamformer according to the attained transmit power. Specifically, we apply a quadratic function as the minorizer, leading to a low-complexity solution at each iteration. In addition, to further facilitate low-cost implementation of the constant-envelope beamformer, we consider the problem of one-bit beamforming design and propose an efficient iterative method based on the Nesterov-like gradient method to solve it. Numerical simulations are provided to demonstrate the effectiveness of the proposed schemes.