Before deploying a language model (LM) within a given domain, it is important to measure its tendency to generate factually incorrect information in that domain. Existing factual generation evaluation methods focus on facts sampled from the LM itself, and thus do not control the set of evaluated facts and might under-represent rare and unlikely facts. We propose FACTOR: Factual Assessment via Corpus TransfORmation, a scalable approach for evaluating LM factuality. FACTOR automatically transforms a factual corpus of interest into a benchmark evaluating an LM's propensity to generate true facts from the corpus vs. similar but incorrect statements. We use our framework to create two benchmarks: Wiki-FACTOR and News-FACTOR. We show that: (i) our benchmark scores increase with model size and improve when the LM is augmented with retrieval; (ii) benchmark score correlates with perplexity, but the two metrics do not always agree on model ranking; and (iii) when perplexity and benchmark score disagree, the latter better reflects factuality in open-ended generation, as measured by human annotators. We make our data and code publicly available in //github.com/AI21Labs/factor.
相關內容
With their recent development, large language models (LLMs) have been found to exhibit a certain level of Theory of Mind (ToM), a complex cognitive capacity that is related to our conscious mind and that allows us to infer another's beliefs and perspective. While human ToM capabilities are believed to derive from the neural activity of a broadly interconnected brain network, including that of dorsal medial prefrontal cortex (dmPFC) neurons, the precise processes underlying LLM's capacity for ToM or their similarities with that of humans remains largely unknown. In this study, we drew inspiration from the dmPFC neurons subserving human ToM and employed a similar methodology to examine whether LLMs exhibit comparable characteristics. Surprisingly, our analysis revealed a striking resemblance between the two, as hidden embeddings (artificial neurons) within LLMs started to exhibit significant responsiveness to either true- or false-belief trials, suggesting their ability to represent another's perspective. These artificial embedding responses were closely correlated with the LLMs' performance during the ToM tasks, a property that was dependent on the size of the models. Further, the other's beliefs could be accurately decoded using the entire embeddings, indicating the presence of the embeddings' ToM capability at the population level. Together, our findings revealed an emergent property of LLMs' embeddings that modified their activities in response to ToM features, offering initial evidence of a parallel between the artificial model and neurons in the human brain.
The advanced language processing abilities of large language models (LLMs) have stimulated debate over their capacity to replicate human-like cognitive processes. One differentiating factor between language processing in LLMs and humans is that language input is often grounded in several perceptual modalities, whereas most LLMs process solely text-based information. Multimodal grounding allows humans to integrate - e.g. visual context with linguistic information and thereby place constraints on the space of upcoming words, reducing cognitive load and improving comprehension. Recent multimodal LLMs (mLLMs) combine a visual-linguistic embedding space with a transformer type attention mechanism for next-word prediction. Here we ask whether predictive language processing based on multimodal input in mLLMs aligns with humans. Two-hundred participants watched short audio-visual clips and estimated predictability of an upcoming verb or noun. The same clips were processed by the mLLM CLIP, with predictability scores based on comparing image and text feature vectors. Eye-tracking was used to estimate what visual features participants attended to, and CLIP's visual attention weights were recorded. We find that alignment of predictability scores was driven by multimodality of CLIP (no alignment for a unimodal state-of-the-art LLM) and by the attention mechanism (no alignment when attention weights were perturbated or when the same input was fed to a multimodal model without attention). We further find a significant spatial overlap between CLIP's visual attention weights and human eye-tracking data. Results suggest that comparable processes of integrating multimodal information, guided by attention to relevant visual features, supports predictive language processing in mLLMs and humans.
The way we analyse clinical texts has undergone major changes over the last years. The introduction of language models such as BERT led to adaptations for the (bio)medical domain like PubMedBERT and ClinicalBERT. These models rely on large databases of archived medical documents. While performing well in terms of accuracy, both the lack of interpretability and limitations to transfer across languages limit their use in clinical setting. We introduce a novel light-weight graph-based embedding method specifically catering radiology reports. It takes into account the structure and composition of the report, while also connecting medical terms in the report through the multi-lingual SNOMED Clinical Terms knowledge base. The resulting graph embedding uncovers the underlying relationships among clinical terms, achieving a representation that is better understandable for clinicians and clinically more accurate, without reliance on large pre-training datasets. We show the use of this embedding on two tasks namely disease classification of X-ray reports and image classification. For disease classification our model is competitive with its BERT-based counterparts, while being magnitudes smaller in size and training data requirements. For image classification, we show the effectiveness of the graph embedding leveraging cross-modal knowledge transfer and show how this method is usable across different languages.
Topic modeling and text mining are subsets of Natural Language Processing with relevance for conducting meta-analysis (MA) and systematic review (SR). For evidence synthesis, the above NLP methods are conventionally used for topic-specific literature searches or extracting values from reports to automate essential phases of SR and MA. Instead, this work proposes a comparative topic modeling approach to analyze reports of contradictory results on the same general research question. Specifically, the objective is to find topics exhibiting distinct associations with significant results for an outcome of interest by ranking them according to their proportional occurrence and consistency of distribution across reports of significant results. The proposed method was tested on broad-scope studies addressing whether supplemental nutritional compounds significantly benefit macular degeneration (MD). Eight compounds were identified as having a particular association with reports of significant results for benefitting MD. Six of these were further supported in terms of effectiveness upon conducting a follow-up literature search for validation (omega-3 fatty acids, copper, zeaxanthin, lutein, zinc, and nitrates). The two not supported by the follow-up literature search (niacin and molybdenum) also had the lowest scores under the proposed methods ranking system, suggesting that the proposed method's score for a given topic is a viable proxy for its degree of association with the outcome of interest. These results underpin the proposed methods potential to add specificity in understanding effects from broad-scope reports, elucidate topics of interest for future research, and guide evidence synthesis in a systematic and scalable way.
Ever since the emergence of large language models (LLMs) and related applications, such as ChatGPT, its performance and error analysis for programming tasks have been subject to research. In this work-in-progress paper, we explore the potential of such LLMs for computing educators and learners, as we analyze the feedback it generates to a given input containing program code. In particular, we aim at (1) exploring how an LLM like ChatGPT responds to students seeking help with their introductory programming tasks, and (2) identifying feedback types in its responses. To achieve these goals, we used students' programming sequences from a dataset gathered within a CS1 course as input for ChatGPT along with questions required to elicit feedback and correct solutions. The results show that ChatGPT performs reasonably well for some of the introductory programming tasks and student errors, which means that students can potentially benefit. However, educators should provide guidance on how to use the provided feedback, as it can contain misleading information for novices.
As artificial intelligence (AI) models continue to scale up, they are becoming more capable and integrated into various forms of decision-making systems. For models involved in moral decision-making, also known as artificial moral agents (AMA), interpretability provides a way to trust and understand the agent's internal reasoning mechanisms for effective use and error correction. In this paper, we provide an overview of this rapidly-evolving sub-field of AI interpretability, introduce the concept of the Minimum Level of Interpretability (MLI) and recommend an MLI for various types of agents, to aid their safe deployment in real-world settings.
Large language models (LLMs) have significantly advanced the field of natural language processing (NLP), providing a highly useful, task-agnostic foundation for a wide range of applications. The great promise of LLMs as general task solvers motivated people to extend their functionality largely beyond just a ``chatbot'', and use it as an assistant or even replacement for domain experts and tools in specific domains such as healthcare, finance, and education. However, directly applying LLMs to solve sophisticated problems in specific domains meets many hurdles, caused by the heterogeneity of domain data, the sophistication of domain knowledge, the uniqueness of domain objectives, and the diversity of the constraints (e.g., various social norms, cultural conformity, religious beliefs, and ethical standards in the domain applications). To fill such a gap, explosively-increase research, and practices have been conducted in very recent years on the domain specialization of LLMs, which, however, calls for a comprehensive and systematic review to better summarizes and guide this promising domain. In this survey paper, first, we propose a systematic taxonomy that categorizes the LLM domain-specialization techniques based on the accessibility to LLMs and summarizes the framework for all the subcategories as well as their relations and differences to each other. We also present a comprehensive taxonomy of critical application domains that can benefit from specialized LLMs, discussing their practical significance and open challenges. Furthermore, we offer insights into the current research status and future trends in this area.
Graph neural networks (GNNs) have been demonstrated to be a powerful algorithmic model in broad application fields for their effectiveness in learning over graphs. To scale GNN training up for large-scale and ever-growing graphs, the most promising solution is distributed training which distributes the workload of training across multiple computing nodes. However, the workflows, computational patterns, communication patterns, and optimization techniques of distributed GNN training remain preliminarily understood. In this paper, we provide a comprehensive survey of distributed GNN training by investigating various optimization techniques used in distributed GNN training. First, distributed GNN training is classified into several categories according to their workflows. In addition, their computational patterns and communication patterns, as well as the optimization techniques proposed by recent work are introduced. Second, the software frameworks and hardware platforms of distributed GNN training are also introduced for a deeper understanding. Third, distributed GNN training is compared with distributed training of deep neural networks, emphasizing the uniqueness of distributed GNN training. Finally, interesting issues and opportunities in this field are discussed.
Knowledge graphs represent factual knowledge about the world as relationships between concepts and are critical for intelligent decision making in enterprise applications. New knowledge is inferred from the existing facts in the knowledge graphs by encoding the concepts and relations into low-dimensional feature vector representations. The most effective representations for this task, called Knowledge Graph Embeddings (KGE), are learned through neural network architectures. Due to their impressive predictive performance, they are increasingly used in high-impact domains like healthcare, finance and education. However, are the black-box KGE models adversarially robust for use in domains with high stakes? This thesis argues that state-of-the-art KGE models are vulnerable to data poisoning attacks, that is, their predictive performance can be degraded by systematically crafted perturbations to the training knowledge graph. To support this argument, two novel data poisoning attacks are proposed that craft input deletions or additions at training time to subvert the learned model's performance at inference time. These adversarial attacks target the task of predicting the missing facts in knowledge graphs using KGE models, and the evaluation shows that the simpler attacks are competitive with or outperform the computationally expensive ones. The thesis contributions not only highlight and provide an opportunity to fix the security vulnerabilities of KGE models, but also help to understand the black-box predictive behaviour of KGE models.
Neural machine translation (NMT) is a deep learning based approach for machine translation, which yields the state-of-the-art translation performance in scenarios where large-scale parallel corpora are available. Although the high-quality and domain-specific translation is crucial in the real world, domain-specific corpora are usually scarce or nonexistent, and thus vanilla NMT performs poorly in such scenarios. Domain adaptation that leverages both out-of-domain parallel corpora as well as monolingual corpora for in-domain translation, is very important for domain-specific translation. In this paper, we give a comprehensive survey of the state-of-the-art domain adaptation techniques for NMT.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191