Recent advancements in knowledge graph question answering (KGQA) have shown promise, yet existing methods often fail to align with human reasoning patterns that involve continuous reflection and refinement. This paper proposes FD-PORT (flow-guided direct preference optimization for knowledge graph reasoning with trees), a novel approach that combines Monte Carlo Tree Search (MCTS) with flow-guided direct preference optimization (FDPO) for KGQA tasks. MCTS simulates human-like reasoning by systematically exploring multiple inference paths in knowledge graphs, while FDPO transforms the search feedback into fine-grained training signals through flow balance conditions. Unlike traditional methods focusing on end-to-end training or sequence-level preferences, FD-PORT establishes flow consistency between any states along the reasoning chain, enabling robust multi-hop reasoning that adapts to local decisions and long-range dependencies. Experimental results on three benchmark datasets demonstrate that FD-PORT significantly outperforms state-of-the-art methods, achieving up to 50.6% improvements over GPT-4 on complex multi-hop reasoning tasks with a smaller open-source language model. The framework is advanced in maintaining diverse reasoning paths while ensuring answer quality, closely mirroring human problem-solving strategies.

Empathetic Response Generation (ERG) is one of the key tasks of the affective computing area, which aims to produce emotionally nuanced and compassionate responses to user's queries. However, existing ERG research is predominantly confined to the singleton text modality, limiting its effectiveness since human emotions are inherently conveyed through multiple modalities. To combat this, we introduce an avatar-based Multimodal ERG (MERG) task, entailing rich text, speech, and facial vision information. We first present a large-scale high-quality benchmark dataset, AvaMERG, which extends traditional text ERG by incorporating authentic human speech audio and dynamic talking-face avatar videos, encompassing a diverse range of avatar profiles and broadly covering various topics of real-world scenarios. Further, we deliberately tailor a system, named Empatheia, for MERG. Built upon a Multimodal Large Language Model (MLLM) with multimodal encoder, speech and avatar generators, Empatheia performs end-to-end MERG, with Chain-of-Empathetic reasoning mechanism integrated for enhanced empathy understanding and reasoning.Finally, we devise a list of empathetic-enhanced tuning strategies, strengthening the capabilities of emotional accuracy and content, avatar-profile consistency across modalities. Experimental results on AvaMERG data demonstrate that Empatheia consistently shows superior performance than baseline methods on both textual ERG and MERG. All data and code are open at https://AvaMERG.github.io/.

Logical reasoning consistently plays a fundamental and significant role in the domains of knowledge engineering and artificial intelligence. Recently, Large Language Models (LLMs) have emerged as a noteworthy innovation in natural language processing (NLP). However, the question of whether LLMs can effectively address the task of logical reasoning, which requires gradual cognitive inference similar to human intelligence, remains unanswered. To this end, we aim to bridge this gap and provide comprehensive evaluations in this paper. First, to offer systematic evaluations, we select fifteen typical logical reasoning datasets and organize them into deductive, inductive, abductive and mixed-form reasoning settings. Considering the comprehensiveness of evaluations, we include 3 early-era representative LLMs and 4 trending LLMs. Second, different from previous evaluations relying only on simple metrics (e.g.,accuracy), we propose fine-level evaluations in objective and subjective manners, covering both answers and explanations, includinganswer correctness,explain correctness,explain completenessandexplain redundancy. Additionally, to uncover the logical flaws of LLMs, problematic cases will be attributed to five error types from two dimensions, i.e.,evidence selection processandreasoning process. Third, to avoid the influences of knowledge bias and concentrate purely on benchmarking the logical reasoning capability of LLMs, we propose a new dataset with neutral content. Based on the in-depth evaluations, this paper finally forms a general evaluation scheme of logical reasoning capability from six dimensions (i.e.,Correct,Rigorous,Self-aware,Active,OrientedandNo hallucination). It reflects the pros and cons of LLMs and gives guiding directions for future works.

Predicting stock price movements is a high-stakes task that demands explainability for human decision-makers. A key shortcoming in current methods is treating sub-predictions independently, without learning from accumulated experiences. We propose a novel triplet network for contrastive learning to enhance the explainability of stock movement prediction by considering instances of "integrated textual information and quantitative indicators". We refer to the target past-l-day tweet-price time series as the "anchor instance". Each anchor instance is paired with a "positive instance" characterized by highly correlated return trends yet significant differences across the entire feature space, and a "negative instance" that exhibits similar return trends along with high proximity in the feature space. The model is designed with the objective of (1) minimizing the cross entropy loss between input logits and target, (2) minimizing the distance between the anchor instances and positive instances, and (3) maximizing the distance between the anchor instances and negative instances. Our framework's effectiveness is demonstrated through extensive testing, showing superior performance on stock prediction benchmarks.

Cognitive reasoning holds a significant place within Natural Language Processing (NLP). Yet, the exploration of zero-shot scenarios, which align more closely with real-life situations than supervised scenarios, has been relatively limited. While a few studies have employed Large Language Models (LLMs) to tackle zero-shot cognitive reasoning tasks, they still grapple with two key challenges: 1) Traditional approaches rely on the chain-of-thought (CoT) mechanism, wherein LLMs are provided with a "Let's think step by step'' prompt. However, this schema may not accurately understand the meaning of a given question and ignores the possible learned knowledge (e.g., background or commonsense) of the LLMs about the questions, leading to incorrect answers. 2) Previous CoT methods normally exploit a single Large Language Model (LLM) and design many strategies to augment this LLM. We argue that the power of a single LLM is typically finite since it may not have learned some relevant knowledge about the question. To address these issues, we propose a Multi-LLM Knowledge Fusion (MLKF) approach, which resorts to heterogeneous knowledge emerging from multiple LLMs, for zero-shot cognitive reasoning tasks. Through extensive experiments and detailed analysis, we demonstrate that our MLKF can outperform the existing zero-shot or unsupervised state-of-the-art methods on four kinds of zero-shot tasks: aspect sentiment analysis, named entity recognition, question answering, and mathematical reasoning. Our code is available at https://github.com/trueBatty/MLKF

Metaphors frequently appear in financial news headlines due to their ability to effectively convey complex financial concepts and market trends in a concise and memorable manner. Cognitive scientists have found that metaphors serve as the reflections of human cognition by means of concept mappings. In this work, we aim to analyze the metaphorical expressions and associated cognitive patterns employed by financial analysts in the headlines of financial analysis reports. Such an examination would enhance our comprehension of the cognitive state of financial analysts regarding various financial trends. We employ the latest computational metaphor processing tool, MetaPro to achieve this target by mining metaphors and cognitive patterns from 1,407,328 financial analyst report headlines, spanning the period from 14 February 2009 to 11 June 2020. We analyze the mined concept mappings by different time periods, and market movements, and deliver novel findings in these two dimensions.

Human coders assign standardized medical codes to clinical documents generated during patients’ hospitalization, which is error prone and labor intensive. Automated medical coding approaches have been developed using machine learning methods, such as deep neural networks. Nevertheless, automated medical coding is still challenging because of complex code association, noise in lengthy documents, and the imbalanced class problem. We propose a novel neural network, called the Multitask Balanced and Recalibrated Neural Network, to solve these issues. Significantly, the multitask learning scheme shares the relationship knowledge between different coding branches to capture code association. A recalibrated aggregation module is developed by cascading convolutional blocks to extract high-level semantic features that mitigate the impact of noise in documents. Also, the cascaded structure of the recalibrated module can benefit learning from lengthy notes. To solve the imbalanced class problem, we deploy focal loss to redistribute the attention on low- and high-frequency medical codes. Experimental results show that our proposed model outperforms competitive baselines on a real-world clinical dataset called the Medical Information Mart for Intensive Care (MIMIC-III).

Abstract Medical coding translates professionally written medical reports into standardized codes, which is an essential part of medical information systems and health insurance reimbursement. Manual coding by trained human coders is time-consuming and error-prone. Thus, automated coding algorithms have been developed, building especially on the recent advances in machine learning and deep neural networks. To solve the challenges of encoding lengthy and noisy clinical documents and capturing code associations, we propose a multitask recalibrated aggregation network. In particular, multitask learning shares information across different coding schemes and captures the dependencies between different medical codes. Feature recalibration and aggregation in shared modules enhance representation learning for lengthy notes. Experiments with a real-world MIMIC-III dataset show significantly improved predictive performance.

Deep learning has unlocked new paths towards the emulation of the peculiarly-human capability of learning from examples. While this kind of bottom-up learning works well for tasks such as image classification or object detection, it is not as effective when it comes to natural language processing. Communication is much more than learning a sequence of letters and words: it requires a basic understanding of the world and social norms, cultural awareness, commonsense knowledge, etc.; all things that we mostly learn in a top-down manner. In this work, we integrate top-down and bottom-up learning via an ensemble of symbolic and subsymbolic AI tools, which we apply to the interesting problem of polarity detection from text. In particular, we integrate logical reasoning within deep learning architectures to build a new version of SenticNet, a commonsense knowledge base for sentiment analysis.

State-of-the-art personality prediction with text data mostly relies on bottom up, automated feature generation as part of the deep learning process. More traditional models rely on hand-crafted, theory-based text-feature categories. We propose a novel deep learning-based model which integrates traditional psycholinguistic features with language model embeddings to predict personality from the Essays dataset for Big-Five and Kaggle dataset for MBTI. With this approach we achieve state-of-the-art model performance. Additionally, we use interpretable machine learning to visualize and quantify the impact of various language features in the respective personality prediction models. We conclude with a discussion on the potential this work has for computational modeling and psychological science alike.

Gender-based-violence is a serious concern in recent times. Due to the social stigma attached to these assaults, victims rarely come forward. Implementing policy measures to prevent sexual violence get constrained due to lack of crime statistics. However, the recent outcry on the Twitter platform allows us to address this concern. Sexual assaults occur at workplaces, public places, educational institutes and also at home. Policy level approaches and awareness campaign for these assaults would not be similar. So, we want to identify the risk factor associated with these sexual assaults. We extracted 0.7 million tweets during the #MeToo social media movement. Next, we employ deep learning techniques to classify these sexual violences. We observe that sexual assaults by a family member at own home is a more serious concern than harassment by a stranger at public places. This study reveals assaults by a known person are more prevalent than assaults by unknown strangers.

Information extraction in the medical domain is laborious and time-consuming due to the insufficient number of domainspecific lexicons and lack of involvement of domain experts such as doctors and medical practitioners.Thus, in the present work, we are motivated to design a new lexicon, WME 3.0 (WordNet of Medical Events), which contains over 10,000 medical concepts along with their part of speech, gloss (descriptive explanations), polarity score, sentiment, similar sentiment words, category, affinity score and gravity score features.In addition, the manual annotators help to validate the overall as well as individual category level of medical concepts of WME 3.0 using Cohen's Kappa agreement metric.The agreement score indicates almost correct identification of medical concepts and their assigned features in WME 3.0.

We find from four datasets that time expressions are formed by loose structure and the words used to express time information can differentiate time expressions from common text. The findings drive us to design a learning method named TOMN to model time expressions. TOMN defines a time-related tagging scheme named TOMN scheme with four tags, namely \tomnT,\tomnO, \tomnM,and \tomnN, indicating the constituents of time expression, namely \tomnT ime token, \tomnM odifier, \tomnN umeral, and the words \tomnO utside time expression. In modeling, TOMN assigns a word with a TOMN tag under conditional random fields with minimal features. Essentially, our constituent-based TOMN scheme overcomes the problem of inconsistent tag assignment that is caused by the conventional position-based tagging schemes (\eg BIO scheme and BILOU scheme). Experiments show that TOMN is equally or more effective than state-of-the-art methods on various datasets, and much more robust on cross-datasets. Moreover, our analysis can explain many empirical observations in other works about time expression recognition and named entity recognition.

In this paper, we study an important yet largely under-explored setting of graph embedding, i.e., embedding communities instead of each individual nodes. We find that community embedding is not only useful for community-level applications such as graph visualization, but also beneficial to both community detection and node classification. To learn such embedding, our insight hinges upon a closed loop among community embedding, community detection and node embedding. On the one hand, node embedding can help improve community detection, which outputs good communities for fitting better community embedding. On the other hand, community embedding can be used to optimize the node embedding by introducing a community-aware high-order proximity. Guided by this insight, we propose a novel community embedding framework that jointly solves the three tasks together. We evaluate such a framework on multiple real-world datasets, and show that it improves graph visualization and outperforms state-of-the-art baselines in various application tasks, e.g., community detection and node classification.

Multimodal sentiment analysis involves identifying sentiment in videos and is a developing field of research. Unlike current works, which model utterances individually, we propose a recurrent model that is able to capture contextual information among utterances. In this paper, we also introduce attentionbased networks for improving both context learning and dynamic feature fusion. Our model shows 6-8% improvement over the state of the art on a benchmark dataset.

Technology has enabled anyone with an Internet connection to easily create and share their ideas, opinions and content with millions of other people around the world. Much of the content being posted and consumed online is multimodal. With billions of phones, tablets and PCs shipping today with built-in cameras and a host of new video-equipped wearables like Google Glass on the horizon, the amount of video on the Internet will only continue to increase. It has become increasingly difficult for researchers to keep up with this deluge of multimodal content, let alone organize or make sense of it. Mining useful knowledge from video is a critical need that will grow exponentially, in pace with the global growth of content. This is particularly important in sentiment analysis, as both service and product reviews are gradually shifting from unimodal to multimodal. We present a novel method to extract features from visual and textual modalities using deep convolutional neural networks. By feeding such features to a multiple kernel learning classifier, we significantly outperform the state of the art of multimodal emotion recognition and sentiment analysis on different datasets.

In order to overcome the lack of medical corpora, we have developed a WordNet for Medical Events (WME) for identifying medical terms and their sense related information using a seed list.The initial WME resource contains 1654 medical terms or concepts.In the present research, we have reported the enhancement of WME with 6415 number of medical concepts along with their conceptual features viz.Parts-of-Speech (POS), gloss, semantics, polarity, sense and affinity.Several polarity lexicons viz.SentiWordNet, SenticNet, Bing Liu's subjectivity list and Taboda's adjective list were introduced with WordNet synonyms and hyponyms for expansion.The semantics feature guided us to build a semantic co-reference relation based network between the related medical concepts.These features help to prepare a medical concept network for better sense relation based visualization.Finally, we evaluated with respect to Adaptive Lesk Algorithm and conducted an agreement analysis for validating the expanded WME resource.

The automatic detection of orientation and emotions in texts is becoming increasingly important in the Web 2.0 scenario. There is a considerable need for innovative techniques and tools capable of identifying and detecting the attitude of unstructured text. The paper tackles two crucial aspects of the sentiment classification problem: first, the computational complexity of the deployed framework; second, the ability of the framework itself to operate effectively in heterogeneous commercial domains. The proposed approach adopts empirical learning to implement the sentiment-classification technology, and uses a distance-based predictive model to combine computational efficiency and modularity. A suitably designed semantic-based metric is the cognitive core that measures the distance between two user reviews, according to the sentiment they communicate. The framework ultimately nullifies the training process; at the same time, it takes advantage of a classification procedure whose computational cost increases linearly when the training corpus increases. To attain an objective measurement of the actual accuracy of the sentiment classification method, a campaign of tests involved a pair of complex, real-world scoring domains; the goal was to compare the predicted sentiment scores with actual scores provided by human assessors. Experimental results confirmed that the overall approach attained satisfactory performances in terms of both cross-domain classification accuracy and computational efficiency.