Pair Extraction Research Articles

Multimodal Emotion-Cause Pair Extraction with Holistic Interaction and Label Constraint

Multimodal Emotion-Cause Pair Extraction with Holistic Interaction and Label Constraint

Overview and future prospects of emotion-cause pair extraction

Overview and future prospects of emotion-cause pair extraction

Unifying emotion-oriented and cause-oriented predictions for emotion-cause pair extraction

Emotion-cause pair extraction (ECPE) is an extraction task aiming to simultaneously identify the emotions and causes from the text without emotion annotations. Let ci and cj represent the emotion clause and the cause clause of a document, respectively, and we can predict one from the other and vice versa. Previous works fail to take advantage of this bidirectional opportunity. We refer to the prediction from ci to cj, i.e., ci→cj, as an emotion-oriented cause prediction (EoCP) task and the prediction from cj to ci, i.e., cj→ci, as a cause-oriented emotion prediction (CoEP) task. After redefining the ECPE task, we propose a novel unified architecture for ECPE, which incorporates EoCP and CoEP as cells and unifies them into a single-chain architecture. Additionally, we redefine emotion-cause pair extraction as a closed-loop structure detection problem to alleviate the mismatch between emotion and cause clauses. To enhance the training of the architecture, we provide a procedure for estimating the confidence of the extraction system for its emotion-cause pairs. We demonstrate the superiority of our proposed model through extensive experiments on two public datasets, achieving a new state-of-the-art performance. Furthermore, our method particularly achieves significant improvements in multiple emotion-cause pair extraction.

Integrated network pharmacology and serum metabonomics analysis to explore the potential mechanism of Anemarrhena asphodeloides Bunge-Phellodendron chinense Schneid herb pair in the treatment of benign prostatic hyperplasia

Anemarrhena asphodeloides Bunge-Phellodendron chinense Schneid (AAPC) is one of the most widely accepted herb pairs in Chinese medicine prescription for treating benign prostatic hyperplasia (BPH). However, the mechanisms underlying the combination of the two herbs for anti-BPH are still not completely clear. To uncover the potential mechanism of the AAPC herb pair in the treatment of BPH, chemical profiling, network pharmacology, serum metabonomics and experimental validation were integrated. UHPLC-Q-Exactive Orbitrap-MS was performed to characterize the chemical profiling of the herb pair extract, and network pharmacology was employed to forecast the potential effective components, core targets and key signaling pathways. Then, western blot and RT-PCR experiments were conducted to verify the PI3K/Akt/NF-κB signaling pathway predicted by network pharmacology. Finally, the serum differential metabolites and metabolic pathways were analyzed by serum non-targeted metabonomics, and these results were jointly analyzed by MetScape. 51 chemical components of the AAPC herb pair extract were identified, including phellodendrine, magnoflorine, berberine, mangiferin, anemarsaponin BIII, etc. In network pharmacology, the predicted core targets of these components include AKT1, TNF, EGFR, PTGS2, PIK3CA, etc. The KEGG pathway enrichment analysis indicated that PI3K-Akt, Rap1 and MAPK signaling pathways may play a key role in the AAPC herb pair for the treatment of BPH, and the results of animal experiments demonstrated that the herb pair could significantly inhibit the activation and expression of p-PI3K/PI3K, p-Akt/Akt, p-NF-κB/NF-κB in protein and mRNA levels. Furthermore, 31 serum differential metabolites and three main metabolic pathways were obtained by serum non-targeted metabonomics. And the crucial metabolic pathway of arachidonic acid (AA) was obtained by integrated analysis of network pharmacology and metabonomics results. In conclusion, the AAPC herb pair can improve BPH through inhibiting the activation and expression of the PI3K/Akt/NF-κB signaling pathway and AA metabolism.

Ion pair sites for efficient electrochemical extraction of uranium in real nuclear wastewater

Electrochemical uranium extraction from nuclear wastewater represents an emerging strategy for recycling uranium resources. However, in nuclear fuel production which generates the majority of uranium-containing nuclear wastewater, fluoride ion (F−) co-exists with uranyl (UO22+), resulting in the complex species of UO2Fx and thus decreasing extraction efficiency. Herein, we construct Tiδ+-PO43− ion pair extraction sites in Ti(OH)PO4 for efficient electrochemical uranium extraction in wastewater from nuclear fuel production. These sites selectively bind with UO2Fx through the combined Ti-F and multiple O-U-O bonds. In the uranium extraction, the uranium species undergo a crystalline transition from U3O7 to K3UO2F5. In real nuclear wastewater, the uranium is electrochemically extracted with a high efficiency of 99.6% and finally purified as uranium oxide powder, corresponding to an extraction capacity of 6829 mg g−1 without saturation. This work paves an efficient way for electrochemical uranium recycling in real wastewater of nuclear production.

Investigation of the drug–drug interaction and incompatibility mechanism between Aconitum carmichaelii Debx and Pinellia ternata (Thunb.) Breit

Ethnopharmacological relevanceThe combination of Aconitum carmichaelii Debx (Chuanwu, CW) and Pinellia ternata (Thunb.) Breit (Banxia, BX) forms an herbal pair within the eighteen incompatible medicaments (EIM), indicating that BX and CW are incompatible. However, the scientific understanding of this incompatibility mechanism, especially the corresponding drug–drug interaction (DDI), remains complex and unclear. Aim of the studyThis study aims to explain the DDI and potential incompatibility mechanism between CW and BX based on pharmacokinetics and cocktail approach. Materials and methodsUltraperformance liquid chromatography–tandem mass spectrometry methods were established for pharmacokinetics and cocktail studies. To explore the DDI between BX and CW, in the pharmacokinetics study, 10 compounds were determined in rat plasma after administering CW and BX–CW herbal pair extracts. In the cocktail assay, the pharmacokinetic parameters of five probe substrates were utilized to assess the influence of BX on cytochrome P450 (CYP) isoenzyme (dapsone for CYP3A4, phenacetin for CYP1A2, dextromethorphan for CYP2D6, tolbutamide for CYP2C9, and omeprazole for CYP2C19). Finally, the DDI and incompatibility mechanism of CW and BX were integrated to explain the rationality of EIM theory. ResultsBX not only enhances the absorption of aconitine and benzoylaconine but also accelerates the metabolism of mesaconitine, benzoylmesaconine, songorine, and fuziline. Moreover, BX affects the activity of CYP enzymes, which regulate the metabolism of toxic compounds. ConclusionsBX altered the activity of CYP enzymes, consequently affecting the metabolism of toxic compounds from CW. This incompatibility mechanism may be related to the increased absorption of these toxic compounds in vivo.

Improved performance of inorganic CsPbI3 perovskite solar cells with WO3/C60 UTL bilayer as an ETL structure: a computational study

CsPbI3, also known as cesium lead iodide, has garnered significant attention as a potential absorber in perovskite solar cells (PSCs). CsPbI3-PSCs have not matched the high performance of hybrid PSCs. This study aimed to identify an effective combination of charge transport layers. Six-hole transporting layers (HTLs) including Spiro-OMeTAD, Cu2O, CuO, CuAlO2, CuSbS2, and SrCu2O2, as well as five electron transporting layers (ETLs) such as TiO2, WO3, ZnO, IGZO, and CdZnS, were tested separately in 30 PSCs. The findings of this research indicate that CuAlO2 as the HTL and WO3 as the ETL that are the most appropriate materials among the options examined, so we use FTO/WO3/CsPbI3/CuAlO2/Au as a required PSC. In this research, we used SCAPS (Solar Cell Capacitance Simulator)−1D device modeling to investigate the bilayer ETL of inorganic CsPbI3-PSC and discover the methods to improve their efficiency. In planar PSCs, optimizing electron–hole pair extraction and recombination at the ETL/perovskite interface is crucial for achieving high performance. The key concept is to enhance the WO3/perovskite interface properties by adding a 5 nm ultra-thin layer (UTL) of C60. The bilayer structure WO3/C60 was found to have the advantage of high electron extraction and low interfacial recombination, primarily due to more effective energy level alignment and defect passivation. To achieve the superior efficiency of PSC, various factors such as defect and doping densities in all layers, the energy level alteration of ETL and HTL, interface defect densities on both ETL and HTL sides, back metal contact, operating temperature, and parasitic resistances were optimized. After optimizing these parameters, the efficiency of the system containing WO3/C60 bilayer ETL was found to be 29.39%. The current work proposes a straightforward and promising method to create photovoltaic devices, especially for many types of perovskites, with desirable charge transport layers and recombination properties.

MV-SHIF: Multi-view symmetric hypothesis inference fusion network for emotion-cause pair extraction in documents

Emotion-cause pair extraction (ECPE) is a challenging task that aims to automatically identify pairs of emotions and their causes from documents. The difficulty of ECPE lies in distinguishing valid emotion-cause pairs from many irrelevant ones. Most previous methods have primarily focused on utilizing multi-task learning to extract semantic information solely from documents without explicitly encoding the relations between clauses. We propose a new approach that incorporates textual entailment paradigm aiming to infer the entailment relationship between the original document as the premise and the clauses or pairs described as the hypothesis. Our approach designs label-view hypothesis templates to improve ECPE by filtering out irrelevant emotion and cause clauses. Furthermore, we formulate candidate emotion-cause pairs as hypothesis statements, and define explicit multi-view symmetric templates to capture the emotion-cause relation semantics. The text entailment recognition for ECPE is finally implemented by fusing multi-view semantic information using a simplified capsule network. Our proposed model achieves state-of-the-art performance on ECPE compared to previous baselines. More importantly, this work demonstrates a novel effective way of applying the textual entailment paradigm to ECPE or clause-level causal discovery by designing multi-view hypothesis inference and information fusion.

From Coarse to Fine: A Distillation Method for Fine-Grained Emotion-Causal Span Pair Extraction in Conversation

We study the problem of extracting emotions and the causes behind these emotions in conversations. Existing methods either tackle them separately or jointly model them at the coarse-grained level of emotions (fewer emotion categories) and causes (utterance-level causes). In this work, we aim to jointly extract more fine-grained emotions and causes. We construct a fine-grained dataset FG-RECCON, includes 16 fine-grained emotion categories and span-level causes. To further improve the fine-grained extraction performance, we propose to utilize the casual discourse knowledge in a knowledge distillation way. Specifically, the teacher model learns to predict causal connective words between utterances, and then guides the student model in identifying both the fine-grained emotion labels and causal spans. Experimental results demonstrate that our distillation method achieves state-of-the-art performance on both RECCON and FG-RECCON dataset.

Oscillating photonic Bell state from a semiconductor quantum dot for quantum key distribution

An on-demand source of bright entangled photon pairs is desirable for quantum key distribution (QKD) and quantum repeaters. The leading candidate to generate such pairs is based on spontaneous parametric down-conversion (SPDC) in non-linear crystals. However, its pair extraction efficiency is limited to 0.1% when operating at near-unity fidelity due to multiphoton emission at high brightness. Quantum dots in photonic nanostructures can in principle overcome this limit, but the devices with high entanglement fidelity (99%) have low pair extraction efficiency (0.01%). Here, we show a measured peak entanglement fidelity of 97.5% ± 0.8% and pair extraction efficiency of 0.65% from an InAsP quantum dot in an InP photonic nanowire waveguide. We show that the generated oscillating two-photon Bell state can establish a secure key for peer-to-peer QKD. Using our time-resolved QKD scheme alleviates the need to remove the quantum dot energy splitting of the intermediate exciton states in the biexciton-exciton cascade.

Emotion-cause pair extraction via knowledge-driven multi-classification and graph-based position embedding

Emotion-cause pair extraction via knowledge-driven multi-classification and graph-based position embedding

Development of target-based cell membrane affinity ultrafiltration technology for a simplified approach to discovering potential bioactive compounds in natural products.

Target-based drug discovery technology based on cell membrane targets has gained significant traction and has been steadily advancing. However, current methods still face certain limitations that need to be addressed. One of the challenges is the laborious preparation process of screening materials, which can be time-consuming and resource-intensive. Additionally, there is a potential issue of non-specific adsorption caused by carrier materials, which can result in false-positive results and compromise the accuracy of the screening process. To address these challenges, this paper proposes a target-based cell membrane affinity ultrafiltration technology for active ingredient discovery in natural products. In this technique, the cell membranes of human lung adenocarcinoma epithelial cells (A549) with a high expression of epidermal growth factor receptor (EGFR) were incubated with candidate drugs and then transferred to an ultrafiltration tube. Through centrifugation, components that interacted with EGFR were retained in the ultrafiltration tube as "EGFR-ligand" complex, while the components that did not interact with EGFR were separated. After thorough washing and eluting, the components interacting with EGFR were dissociated and further identified using LC-MS, enabling the discovery of bioactive compounds. Moreover, the target-based cell membrane affinity ultrafiltration technology exhibited commendable binding capacity and selectivity. Ultimately, this technology successfully screened and identified two major components from the Curcumae Rhizoma-Sparganii Rhizoma (CS) herb pair extracts, which were further validated for their potential anti-tumor activity through pharmacological experiments. By eliminating the need for laborious preparation of screening materials and the potential non-specific adsorption caused by carriers, the development of target-based cell membrane affinity ultrafiltration technology provides a simplified approach and method for bioactive compounds discovery in natural sources.

A machine reading comprehension model with counterfactual contrastive learning for emotion-cause pair extraction

A machine reading comprehension model with counterfactual contrastive learning for emotion-cause pair extraction

A knowledge-guided graph attention network for emotion-cause pair extraction

Emotion-Cause Pair Extraction (ECPE) is a research objective focused on identifying and extracting all emotion-clause and cause-clause pairs from unannotated emotional text. Traditional methodologies have predominantly employed attention mechanisms or joint learning techniques for feature information interaction. However, these approaches often overlook the aggregation of features under the guidance of external knowledge. To enhance performance in addressing this ECPE challenge, we propose a novel knowledge-guided graph attention network, i.e., GAT-ECPE. This model chiefly relies on an interclause dependency graph as a guiding principle. By employing this knowledge-guided graph attention network, we can proficiently combine semantic and structural information between clauses. In addition, an interpair possibility graph, derived from the outcomes of subtasks, is integrated as an additional guiding principle. As such, we are able to aggregate features between clause pairs, thereby facilitating interaction between multiple tasks. Extensive experiments were conducted to validate our proposed model, and the obtained results demonstrate its superiority when compared to 12 considered baselines. In terms of performance metrics, our model achieves an F1 score F1 of 74.92% and a recall R of 77.52%. These values significantly outperform those achieved by state-of-the-art approaches, indicating the effectiveness and superiority of our GAT-ECPE.

Ammonium halide selective ion pair recognition and extraction with a chalcogen bonding heteroditopic receptor.

The first example of a heteroditopic receptor capable of cooperative recognition and extraction of ammonium salt (NH4X) ion-pairs is described. Consisting of a bidentate 3,5-bis-tellurotriazole chalcogen bond donor binding cleft, the appendage of benzo-15-crown-5 (B15C5) substituents to the tellurium centres facilitates binding of the ammonium cation via a co-facial bis-B15C5 sandwich complex, which serves to switch on chalcogen bonding-mediated anion binding potency. Extensive quantitative ion-pair recognition 1H NMR titration studies in CD3CN/CDCl3 (1 : 1, v/v) solvent media reveal impressive ion-pair binding affinities towards a variety of ammonium halide, nitrate and thiocyanate salts, with the heteroditopic receptor displaying notable ammonium halide salt selectivity. The prodigious solution phase NH4X recognition also translates to efficient solid-liquid and liquid-liquid extraction capabilities.

Improving Representation With Hierarchical Contrastive Learning for Emotion-Cause Pair Extraction

Improving Representation With Hierarchical Contrastive Learning for Emotion-Cause Pair Extraction

Efficacy of 4% articaine with 1:100000 epinephrine versus 2% lignocaine with 1:100000 epinephrine in infiltration for extraction of maxillary premolars for orthodontic correction – A prospective, randomized, double-blind, crossover study

Background: Unlike other amide local anesthetics (LAs), articaine has better lipid solubility and potency. Very few studies have been done on comparing articaine with other LAs as buccal infiltration among the participants requiring paired extraction of maxillary premolars but limitations of those studies include smaller sample size, lack of washout periods, and diversity in study designs. Aims and Objectives: This study compares the anesthetic efficacy of 4% articaine (1:100,000 epinephrine) with 2% lidocaine (1:100,000 epinephrine) as buccal infiltration. Materials and Methods: This is a randomized double-blinded crossover study conducted at the Department of Oral and Maxillofacial Surgery in collaboration with the Department of Pharmacology, SRIHER, Chennai. This study was done on 59 participants on an outpatient basis. The selected participants received 4% articaine or lignocaine with epinephrine (1:100000) in the mucobuccal fold at the maxillary area during their first visit and the same individual received the other local anesthetic drug on the other side in the same area during the second visit, i.e., 1 week from the date of the first visit. All the statistics were calculated using the software Statistical Package for the Social Sciences 16 version. For all primary and secondary outcomes in the study, the Mann–Whitney U-test was employed to compute the mean and standard deviation. Results: The total mean age of the participants included was 21.69. The volume of drug required was 1.578 ± 0.2760 mL for articaine and 2.2068 ± 0.0984 ml for lignocaine. The onset time was 30–45 s with articaine and 75–90 s with lignocaine. A decrease in pain scores was seen in the articaine group compared to the lignocaine group. Conclusion: This study showed better efficacy of articaine in comparison to lignocaine. Articaine may replace lignocaine in orthodontic corrections with its higher efficacy, faster action, and longer duration.

Hierarchical Shared Encoder with Task-specific Transformer Layer Selection for Emotion-Cause Pair Extraction

Hierarchical Shared Encoder with Task-specific Transformer Layer Selection for Emotion-Cause Pair Extraction

Seeing both sides: context-aware heterogeneous graph matching networks for extracting-related arguments

Our research focuses on extracting exchanged views from dialogical documents through argument pair extraction (APE). The objective of this process is to facilitate comprehension of complex argumentative discourse by finding the related arguments. The APE comprises two stages: argument mining and argument matching. Researchers typically employ sequence labeling models for mining arguments and text matching models to calculate the relationships between them, thereby generating argument pairs. However, these approaches fail to capture long-distance contextual information and struggle to fully comprehend the complex structure of arguments. In our work, we propose the context-aware heterogeneous graph matching (HGMN) model for the APE task. First, we design a graph schema specifically tailored to argumentative texts, along with a heterogeneous graph attention network that effectively captures context information and structural information of arguments. Moreover, the text matching between arguments is converted into a graph matching paradigm and a multi-granularity graph matching model is proposed to handle the intricate relationships between arguments at various levels of granularity. In this way, the semantics of argument are modeled structurally and thus capture the complicated correlations between arguments. Extensive experiments are conducted to evaluate the HGMN model, including comparisons with existing methods and the GPT series of large language models (LLM). The results demonstrate that HGMN outperforms the state-of-the-art method.

A mutually enhanced multi-scale relation-aware graph convolutional network for argument pair extraction

A mutually enhanced multi-scale relation-aware graph convolutional network for argument pair extraction