Amit P. Sheth

Vipula Rawte, Swagata Chakraborty, Agnibh Pathak et al. · apple-ml, huggingface

The recent advancements in Large Language Models (LLMs) have garnered widespread acclaim for their remarkable emerging capabilities. However, the issue of hallucination has parallelly emerged as a by-product, posing significant concerns. While some recent endeavors have been made to identify and mitigate different types of hallucination, there has been a limited emphasis on the nuanced categorization of hallucination and associated mitigation methods. To address this gap, we offer a fine-grained discourse on profiling hallucination based on its degree, orientation, and category, along with offering strategies for alleviation. As such, we define two overarching orientations of hallucination: (i) factual mirage (FM) and (ii) silver lining (SL). To provide a more comprehensive understanding, both orientations are further sub-categorized into intrinsic and extrinsic, with three degrees of severity - (i) mild, (ii) moderate, and (iii) alarming. We also meticulously categorize hallucination into six types: (i) acronym ambiguity, (ii) numeric nuisance, (iii) generated golem, (iv) virtual voice, (v) geographic erratum, and (vi) time wrap. Furthermore, we curate HallucInation eLiciTation (HILT), a publicly available dataset comprising of 75,000 samples generated using 15 contemporary LLMs along with human annotations for the aforementioned categories. Finally, to establish a method for quantifying and to offer a comparative spectrum that allows us to evaluate and rank LLMs based on their vulnerability to producing hallucinations, we propose Hallucination Vulnerability Index (HVI). We firmly believe that HVI holds significant value as a tool for the wider NLP community, with the potential to serve as a rubric in AI-related policy-making. In conclusion, we propose two solution strategies for mitigating hallucinations.

Megha Chakraborty, S. M Towhidul Islam Tonmoy, S M Mehedi Zaman et al. · apple-ml, stanford

With the rise of prolific ChatGPT, the risk and consequences of AI-generated text has increased alarmingly. To address the inevitable question of ownership attribution for AI-generated artifacts, the US Copyright Office released a statement stating that 'If a work's traditional elements of authorship were produced by a machine, the work lacks human authorship and the Office will not register it'. Furthermore, both the US and the EU governments have recently drafted their initial proposals regarding the regulatory framework for AI. Given this cynosural spotlight on generative AI, AI-generated text detection (AGTD) has emerged as a topic that has already received immediate attention in research, with some initial methods having been proposed, soon followed by emergence of techniques to bypass detection. This paper introduces the Counter Turing Test (CT^2), a benchmark consisting of techniques aiming to offer a comprehensive evaluation of the robustness of existing AGTD techniques. Our empirical findings unequivocally highlight the fragility of the proposed AGTD methods under scrutiny. Amidst the extensive deliberations on policy-making for regulating AI development, it is of utmost importance to assess the detectability of content generated by LLMs. Thus, to establish a quantifiable spectrum facilitating the evaluation and ranking of LLMs according to their detectability levels, we propose the AI Detectability Index (ADI). We conduct a thorough examination of 15 contemporary LLMs, empirically demonstrating that larger LLMs tend to have a higher ADI, indicating they are less detectable compared to smaller LLMs. We firmly believe that ADI holds significant value as a tool for the wider NLP community, with the potential to serve as a rubric in AI-related policy-making.

Thilini Wijesiriwardene, Vinh Nguyen, Goonmeet Bajaj et al.

The UMLS Metathesaurus integrates more than 200 biomedical source vocabularies. During the Metathesaurus construction process, synonymous terms are clustered into concepts by human editors, assisted by lexical similarity algorithms. This process is error-prone and time-consuming. Recently, a deep learning model (LexLM) has been developed for the UMLS Vocabulary Alignment (UVA) task. This work introduces UBERT, a BERT-based language model, pretrained on UMLS terms via a supervised Synonymy Prediction (SP) task replacing the original Next Sentence Prediction (NSP) task. The effectiveness of UBERT for UMLS Metathesaurus construction process is evaluated using the UMLS Vocabulary Alignment (UVA) task. We show that UBERT outperforms the LexLM, as well as biomedical BERT-based models. Key to the performance of UBERT are the synonymy prediction task specifically developed for UBERT, the tight alignment of training data to the UVA task, and the similarity of the models used for pretrained UBERT.

Ishwar B Balappanawar, Ashmit Chamoli, Ruwan Wickramarachchi et al.

Distracted driving is a leading cause of road accidents globally. Identification of distracted driving involves reliably detecting and classifying various forms of driver distraction (e.g., texting, eating, or using in-car devices) from in-vehicle camera feeds to enhance road safety. This task is challenging due to the need for robust models that can generalize to a diverse set of driver behaviors without requiring extensive annotated datasets. In this paper, we propose KiD3, a novel method for distracted driver detection (DDD) by infusing auxiliary knowledge about semantic relations between entities in a scene and the structural configuration of the driver's pose. Specifically, we construct a unified framework that integrates the scene graphs, and driver pose information with the visual cues in video frames to create a holistic representation of the driver's actions.Our results indicate that KiD3 achieves a 13.64% accuracy improvement over the vision-only baseline by incorporating such auxiliary knowledge with visual information.

Deepa Tilwani, Revathy Venkataramanan, Amit P. Sheth

Attribution in large language models (LLMs) remains a significant challenge, particularly in ensuring the factual accuracy and reliability of the generated outputs. Current methods for citation or attribution, such as those employed by tools like Perplexity.ai and Bing Search-integrated LLMs, attempt to ground responses by providing real-time search results and citations. However, so far, these approaches suffer from issues such as hallucinations, biases, surface-level relevance matching, and the complexity of managing vast, unfiltered knowledge sources. While tools like Perplexity.ai dynamically integrate web-based information and citations, they often rely on inconsistent sources such as blog posts or unreliable sources, which limits their overall reliability. We present that these challenges can be mitigated by integrating Neurosymbolic AI (NesyAI), which combines the strengths of neural networks with structured symbolic reasoning. NesyAI offers transparent, interpretable, and dynamic reasoning processes, addressing the limitations of current attribution methods by incorporating structured symbolic knowledge with flexible, neural-based learning. This paper explores how NesyAI frameworks can enhance existing attribution models, offering more reliable, interpretable, and adaptable systems for LLMs.

Vipula Rawte, Sarthak Jain, Aarush Sinha et al.

Recent advances in Large Multimodal Models (LMMs) have expanded their capabilities to video understanding, with Text-to-Video (T2V) models excelling in generating videos from textual prompts. However, they still frequently produce hallucinated content, revealing AI-generated inconsistencies. We introduce ViBe (https://vibe-t2v-bench.github.io/): a large-scale dataset of hallucinated videos from open-source T2V models. We identify five major hallucination types: Vanishing Subject, Omission Error, Numeric Variability, Subject Dysmorphia, and Visual Incongruity. Using ten T2V models, we generated and manually annotated 3,782 videos from 837 diverse MS COCO captions. Our proposed benchmark includes a dataset of hallucinated videos and a classification framework using video embeddings. ViBe serves as a critical resource for evaluating T2V reliability and advancing hallucination detection. We establish classification as a baseline, with the TimeSFormer + CNN ensemble achieving the best performance (0.345 accuracy, 0.342 F1 score). While initial baselines proposed achieve modest accuracy, this highlights the difficulty of automated hallucination detection and the need for improved methods. Our research aims to drive the development of more robust T2V models and evaluate their outputs based on user preferences.

Vipula Rawte, S. M Towhidul Islam Tonmoy, Krishnav Rajbangshi et al. · apple-ml, stanford

The widespread adoption of Large Language Models (LLMs) has facilitated numerous benefits. However, hallucination is a significant concern. In response, Retrieval Augmented Generation (RAG) has emerged as a highly promising paradigm to improve LLM outputs by grounding them in factual information. RAG relies on textual entailment (TE) or similar methods to check if the text produced by LLMs is supported or contradicted, compared to retrieved documents. This paper argues that conventional TE methods are inadequate for spotting hallucinations in content generated by LLMs. For instance, consider a prompt about the 'USA's stance on the Ukraine war''. The AI-generated text states, ...U.S. President Barack Obama says the U.S. will not put troops in Ukraine...'' However, during the war the U.S. president is Joe Biden which contradicts factual reality. Moreover, current TE systems are unable to accurately annotate the given text and identify the exact portion that is contradicted. To address this, we introduces a new type of TE called ``Factual Entailment (FE).'', aims to detect factual inaccuracies in content generated by LLMs while also highlighting the specific text segment that contradicts reality. We present FACTOID (FACTual enTAILment for hallucInation Detection), a benchmark dataset for FE. We propose a multi-task learning (MTL) framework for FE, incorporating state-of-the-art (SoTA) long text embeddings such as e5-mistral-7b-instruct, along with GPT-3, SpanBERT, and RoFormer. The proposed MTL architecture for FE achieves an avg. 40\% improvement in accuracy on the FACTOID benchmark compared to SoTA TE methods. As FE automatically detects hallucinations, we assessed 15 modern LLMs and ranked them using our proposed Auto Hallucination Vulnerability Index (HVI_auto). This index quantifies and offers a comparative scale to evaluate and rank LLMs according to their hallucinations.

Utkarshani Jaimini, Cory Henson, Amit P. Sheth

Causal networks are useful in a wide variety of applications, from medical diagnosis to root-cause analysis in manufacturing. In practice, however, causal networks are often incomplete with missing causal relations. This paper presents a novel approach, called CausalLP, that formulates the issue of incomplete causal networks as a knowledge graph completion problem. More specifically, the task of finding new causal relations in an incomplete causal network is mapped to the task of knowledge graph link prediction. The use of knowledge graphs to represent causal relations enables the integration of external domain knowledge; and as an added complexity, the causal relations have weights representing the strength of the causal association between entities in the knowledge graph. Two primary tasks are supported by CausalLP: causal explanation and causal prediction. An evaluation of this approach uses a benchmark dataset of simulated videos for causal reasoning, CLEVRER-Humans, and compares the performance of multiple knowledge graph embedding algorithms. Two distinct dataset splitting approaches are used for evaluation: (1) random-based split, which is the method typically employed to evaluate link prediction algorithms, and (2) Markov-based split, a novel data split technique that utilizes the Markovian property of causal relations. Results show that using weighted causal relations improves causal link prediction over the baseline without weighted relations.

Vipula Rawte, S. M Towhidul Islam Tonmoy, S M Mehedi Zaman et al. · apple-ml, meta-ai

Hallucination has emerged as the most vulnerable aspect of contemporary Large Language Models (LLMs). In this paper, we introduce the Sorry, Come Again (SCA) prompting, aimed to avoid LLM hallucinations by enhancing comprehension through: (i) optimal paraphrasing and (ii) injecting [PAUSE] tokens to delay LLM generation. First, we provide an in-depth analysis of linguistic nuances: formality, readability, and concreteness of prompts for 21 LLMs, and elucidate how these nuances contribute to hallucinated generation. Prompts with lower readability, formality, or concreteness pose comprehension challenges for LLMs, similar to those faced by humans. In such scenarios, an LLM tends to speculate and generate content based on its imagination (associative memory) to fill these information gaps. Although these speculations may occasionally align with factual information, their accuracy is not assured, often resulting in hallucination. Recent studies reveal that an LLM often neglects the middle sections of extended prompts, a phenomenon termed as lost in the middle. While a specific paraphrase may suit one LLM, the same paraphrased version may elicit a different response from another LLM. Therefore, we propose an optimal paraphrasing technique to identify the most comprehensible paraphrase of a given prompt, evaluated using Integrated Gradient (and its variations) to guarantee that the LLM accurately processes all words. While reading lengthy sentences, humans often pause at various points to better comprehend the meaning read thus far. We have fine-tuned an LLM with injected [PAUSE] tokens, allowing the LLM to pause while reading lengthier prompts. This has brought several key contributions: (i) determining the optimal position to inject [PAUSE], (ii) determining the number of [PAUSE] tokens to be inserted, and (iii) introducing reverse proxy tuning to fine-tune the LLM for [PAUSE] insertion.

Megha Chakraborty, Khushbu Pahwa, Anku Rani et al.

Combating disinformation is one of the burning societal crises -- about 67% of the American population believes that disinformation produces a lot of uncertainty, and 10% of them knowingly propagate disinformation. Evidence shows that disinformation can manipulate democratic processes and public opinion, causing disruption in the share market, panic and anxiety in society, and even death during crises. Therefore, disinformation should be identified promptly and, if possible, mitigated. With approximately 3.2 billion images and 720,000 hours of video shared online daily on social media platforms, scalable detection of multimodal disinformation requires efficient fact verification. Despite progress in automatic text-based fact verification (e.g., FEVER, LIAR), the research community lacks substantial effort in multimodal fact verification. To address this gap, we introduce FACTIFY 3M, a dataset of 3 million samples that pushes the boundaries of the domain of fact verification via a multimodal fake news dataset, in addition to offering explainability through the concept of 5W question-answering. Salient features of the dataset include: (i) textual claims, (ii) ChatGPT-generated paraphrased claims, (iii) associated images, (iv) stable diffusion-generated additional images (i.e., visual paraphrases), (v) pixel-level image heatmap to foster image-text explainability of the claim, (vi) 5W QA pairs, and (vii) adversarial fake news stories.

Mohammad Saeid Mahdavinejad, Mohammadreza Rezvan, Mohammadamin Barekatain et al.

Rapid developments in hardware, software, and communication technologies have allowed the emergence of Internet-connected sensory devices that provide observation and data measurement from the physical world. By 2020, it is estimated that the total number of Internet-connected devices being used will be between 25 and 50 billion. As the numbers grow and technologies become more mature, the volume of data published will increase. Internet-connected devices technology, referred to as Internet of Things (IoT), continues to extend the current Internet by providing connectivity and interaction between the physical and cyber worlds. In addition to increased volume, the IoT generates Big Data characterized by velocity in terms of time and location dependency, with a variety of multiple modalities and varying data quality. Intelligent processing and analysis of this Big Data is the key to developing smart IoT applications. This article assesses the different machine learning methods that deal with the challenges in IoT data by considering smart cities as the main use case. The key contribution of this study is presentation of a taxonomy of machine learning algorithms explaining how different techniques are applied to the data in order to extract higher level information. The potential and challenges of machine learning for IoT data analytics will also be discussed. A use case of applying Support Vector Machine (SVM) on Aarhus Smart City traffic data is presented for a more detailed exploration.

Saeedeh Shekarpour, Faisal Alshargi, Valerie Shalin et al.

While the general analysis of named entities has received substantial research attention on unstructured as well as structured data, the analysis of relations among named entities has received limited focus. In fact, a review of the literature revealed a deficiency in research on the abstract conceptualization required to organize relations. We believe that such an abstract conceptualization can benefit various communities and applications such as natural language processing, information extraction, machine learning, and ontology engineering. In this paper, we present Comprehensive EVent Ontology (CEVO), built on Levin's conceptual hierarchy of English verbs that categorizes verbs with shared meaning, and syntactic behavior. We present the fundamental concepts and requirements for this ontology. Furthermore, we present three use cases employing the CEVO ontology on annotation tasks: (i) annotating relations in plain text, (ii) annotating ontological properties, and (iii) linking textual relations to ontological properties. These use-cases demonstrate the benefits of using CEVO for annotation: (i) annotating English verbs from an abstract conceptualization, (ii) playing the role of an upper ontology for organizing ontological properties, and (iii) facilitating the annotation of text relations using any underlying vocabulary. This resource is available at https://shekarpour.github.io/cevo.io/ using https://w3id.org/cevo namespace.