In natural language generation (NLG), insight mining is seen as a data-to-text task, where data is mined for interesting patterns and verbalised into 'insight' statements. An 'over-generate and rank' paradigm is intuitively used to generate such insights. The multidimensionality and subjectivity of this process make it challenging. This paper introduces a schema-driven method to generate actionable insights from data to drive growth and change. It also introduces a technique to rank the insights to align with user interests based on their feedback. We show preliminary qualitative results of the insights generated using our technique and demonstrate its ability to adapt to feedback.
Prompt engineering is a technique that involves augmenting a large pre-trained model with task-specific hints, known as prompts, to adapt the model to new tasks. Prompts can be created manually as natural language instructions or generated automatically as either natural language instructions or vector representations. Prompt engineering enables the ability to perform predictions based solely on prompts without updating model parameters, and the easier application of large pre-trained models in real-world tasks. In past years, Prompt engineering has been well-studied in natural language processing. Recently, it has also been intensively studied in vision-language modeling. However, there is currently a lack of a systematic overview of prompt engineering on pre-trained vision-language models. This paper aims to provide a comprehensive survey of cutting-edge research in prompt engineering on three types of vision-language models: multimodal-to-text generation models (e.g. Flamingo), image-text matching models (e.g. CLIP), and text-to-image generation models (e.g. Stable Diffusion). For each type of model, a brief model summary, prompting methods, prompting-based applications, and the corresponding responsibility and integrity issues are summarized and discussed. Furthermore, the commonalities and differences between prompting on vision-language models, language models, and vision models are also discussed. The challenges, future directions, and research opportunities are summarized to foster future research on this topic.
Large Language Models (LLMs) have achieved human-level fluency in text generation, making it difficult to distinguish between human-written and LLM-generated texts. This poses a growing risk of misuse of LLMs and demands the development of detectors to identify LLM-generated texts. However, existing detectors lack robustness against attacks: they degrade detection accuracy by simply paraphrasing LLM-generated texts. Furthermore, a malicious user might attempt to deliberately evade the detectors based on detection results, but this has not been assumed in previous studies. In this paper, we propose OUTFOX, a framework that improves the robustness of LLM-generated-text detectors by allowing both the detector and the attacker to consider each other's output. In this framework, the attacker uses the detector's prediction labels as examples for in-context learning and adversarially generates essays that are harder to detect, while the detector uses the adversarially generated essays as examples for in-context learning to learn to detect essays from a strong attacker. Experiments in the domain of student essays show that the proposed detector improves the detection performance on the attacker-generated texts by up to +41.3 points F1-score. Furthermore, the proposed detector shows a state-of-the-art detection performance: up to 96.9 points F1-score, beating existing detectors on non-attacked texts. Finally, the proposed attacker drastically degrades the performance of detectors by up to -57.0 points F1-score, massively outperforming the baseline paraphrasing method for evading detection.
The remarkable capabilities of large-scale language models, such as ChatGPT, in text generation have impressed readers and spurred researchers to devise detectors to mitigate potential risks, including misinformation, phishing, and academic dishonesty. Despite this, most previous studies have been predominantly geared towards creating detectors that differentiate between purely ChatGPT-generated texts and human-authored texts. This approach, however, fails to work on discerning texts generated through human-machine collaboration, such as ChatGPT-polished texts. Addressing this gap, we introduce a novel dataset termed HPPT (ChatGPT-polished academic abstracts), facilitating the construction of more robust detectors. It diverges from extant corpora by comprising pairs of human-written and ChatGPT-polished abstracts instead of purely ChatGPT-generated texts. Additionally, we propose the "Polish Ratio" method, an innovative measure of the degree of modification made by ChatGPT compared to the original human-written text. It provides a mechanism to measure the degree of ChatGPT influence in the resulting text. Our experimental results show our proposed model has better robustness on the HPPT dataset and two existing datasets (HC3 and CDB). Furthermore, the "Polish Ratio" we proposed offers a more comprehensive explanation by quantifying the degree of ChatGPT involvement.
Jina Embeddings constitutes a set of high-performance sentence embedding models adept at translating textual inputs into numerical representations, capturing the semantics of the text. These models excel in applications like dense retrieval and semantic textual similarity. This paper details the development of Jina Embeddings, starting with the creation of high-quality pairwise and triplet datasets. It underlines the crucial role of data cleaning in dataset preparation, offers in-depth insights into the model training process, and concludes with a comprehensive performance evaluation using the Massive Text Embedding Benchmark (MTEB). Furthermore, to increase the model's awareness of grammatical negation, we construct a novel training and evaluation dataset of negated and non-negated statements, which we make publicly available to the community.
This paper investigates how hallucination rates in Large Language Models (LLMs) may be controlled via a symbolic data generation framework, exploring a fundamental relationship between the rate of certain mathematical errors and types of input intervention. Specifically, we systematically generate data for a derivation generation task using a symbolic engine, applying targeted interventions to prompts to perturb features of mathematical derivations such as the surface forms of symbols, equational tree structures, and mathematical context. We then evaluate the effect of prompt interventions across a range of LLMs including fine-tuned T5 models, GPT, and LLaMa-based models. Our experiments suggest that T5-Large can outperform the few-shot performance of GPT-4 on various evaluation sets generated via the framework. However, an extensive evaluation based on human analysis, template-based error detection, and text generation metrics reveals model weaknesses beyond what the reference-based metrics singularly describe. We use these results to tie characteristic distributional footprints of interventions to the human evaluation of LLM derivation quality, potentially leading to significant control over fine-grained mathematical capabilities of language models with respect to specific types of errors.
Since late 2022, Large Language Models (LLMs) have become very prominent with LLMs like ChatGPT and Bard receiving millions of users. Hundreds of new LLMs are announced each week, many of which are deposited to Hugging Face, a repository of machine learning models and datasets. To date, nearly 16,000 Text Generation models have been uploaded to the site. Given the huge influx of LLMs, it is of interest to know which LLM backbones, settings, training methods, and families are popular or trending. However, there is no comprehensive index of LLMs available. We take advantage of the relatively systematic nomenclature of Hugging Face LLMs to perform hierarchical clustering and identify communities amongst LLMs using n-grams and term frequency-inverse document frequency. Our methods successfully identify families of LLMs and accurately cluster LLMs into meaningful subgroups. We present a public web application to navigate and explore Constellation, our atlas of 15,821 LLMs. Constellation rapidly generates a variety of visualizations, namely dendrograms, graphs, word clouds, and scatter plots. Constellation is available at the following link: https://constellation.sites.stanford.edu/.
In this article we show how the problem of neural text generation can be constructively reformulated in terms of transitions between the states of a finite-state machine. This framework leads to an efficient approach to guiding text generation with regular expressions and context-free grammars by allowing the construction of an index over a language model's vocabulary. The approach is model agnostic, allows one to enforce domain-specific knowledge and constraints, and enables the construction of reliable interfaces by guaranteeing the structure of the generated text. It adds little overhead to the token sequence generation process and significantly outperforms existing solutions. An implementation is provided in the open source Python library Outlines
Trustworthiness of generative language models (GLMs) is crucial in their deployment to critical decision making systems. Hence, certified risk control methods such as selective prediction and conformal prediction have been applied to mitigating the hallucination problem in various supervised downstream tasks. However, the lack of appropriate correctness metric hinders applying such principled methods to language generation tasks. In this paper, we circumvent this problem by leveraging the concept of textual entailment to evaluate the correctness of the generated sequence, and propose two selective generation algorithms which control the false discovery rate with respect to the textual entailment relation (FDR-E) with a theoretical guarantee: $\texttt{SGen}^{\texttt{Sup}}$ and $\texttt{SGen}^{\texttt{Semi}}$. $\texttt{SGen}^{\texttt{Sup}}$, a direct modification of the selective prediction, is a supervised learning algorithm which exploits entailment-labeled data, annotated by humans. Since human annotation is costly, we further propose a semi-supervised version, $\texttt{SGen}^{\texttt{Semi}}$, which fully utilizes the unlabeled data by pseudo-labeling, leveraging an entailment set function learned via conformal prediction. Furthermore, $\texttt{SGen}^{\texttt{Semi}}$ enables to use more general class of selection functions, neuro-selection functions, and provides users with an optimal selection function class given multiple candidates. Finally, we demonstrate the efficacy of the $\texttt{SGen}$ family in achieving a desired FDR-E level with comparable selection efficiency to those from baselines on both open and closed source GLMs. Code and datasets are provided at https://github.com/ml-postech/selective-generation.
Large Language Models (LLMs) have demonstrated remarkable abilities across numerous disciplines, primarily assessed through tasks in language generation, knowledge utilization, and complex reasoning. However, their alignment with human emotions and values, which is critical for real-world applications, has not been systematically evaluated. Here, we assessed LLMs' Emotional Intelligence (EI), encompassing emotion recognition, interpretation, and understanding, which is necessary for effective communication and social interactions. Specifically, we first developed a novel psychometric assessment focusing on Emotion Understanding (EU), a core component of EI, suitable for both humans and LLMs. This test requires evaluating complex emotions (e.g., surprised, joyful, puzzled, proud) in realistic scenarios (e.g., despite feeling underperformed, John surprisingly achieved a top score). With a reference frame constructed from over 500 adults, we tested a variety of mainstream LLMs. Most achieved above-average EQ scores, with GPT-4 exceeding 89% of human participants with an EQ of 117. Interestingly, a multivariate pattern analysis revealed that some LLMs apparently did not reply on the human-like mechanism to achieve human-level performance, as their representational patterns were qualitatively distinct from humans. In addition, we discussed the impact of factors such as model size, training method, and architecture on LLMs' EQ. In summary, our study presents one of the first psychometric evaluations of the human-like characteristics of LLMs, which may shed light on the future development of LLMs aiming for both high intellectual and emotional intelligence. Project website: https://emotional-intelligence.github.io/
Current developments in large language models (LLMs) have enabled impressive zero-shot capabilities across various natural language tasks. An interesting application of these systems is in the automated assessment of natural language generation (NLG), a highly challenging area with great practical benefit. In this paper, we explore two options for exploiting the emergent abilities of LLMs for zero-shot NLG assessment: absolute score prediction, and comparative assessment which uses relative comparisons between pairs of candidates. Though comparative assessment has not been extensively studied in NLG assessment, we note that humans often find it more intuitive to compare two options rather than scoring each one independently. This work examines comparative assessment from multiple perspectives: performance compared to absolute grading; positional biases in the prompt; and efficient ranking in terms of the number of comparisons. We illustrate that LLM comparative assessment is a simple, general and effective approach for NLG assessment. For moderate-sized open-source LLMs, such as FlanT5 and Llama2-chat, comparative assessment is superior to prompt scoring, and in many cases can achieve performance competitive with state-of-the-art methods. Additionally, we demonstrate that LLMs often exhibit strong positional biases when making pairwise comparisons, and we propose debiasing methods that can further improve performance.
The recent release of very large language models such as PaLM and GPT-4 has made an unprecedented impact in the popular media and public consciousness, giving rise to a mixture of excitement and fear as to their capabilities and potential uses, and shining a light on natural language processing research which had not previously received so much attention. The developments offer great promise for education technology, and in this paper we look specifically at the potential for incorporating large language models in AI-driven language teaching and assessment systems. We consider several research areas and also discuss the risks and ethical considerations surrounding generative AI in education technology for language learners. Overall we find that larger language models offer improvements over previous models in text generation, opening up routes toward content generation which had not previously been plausible. For text generation they must be prompted carefully and their outputs may need to be reshaped before they are ready for use. For automated grading and grammatical error correction, tasks whose progress is checked on well-known benchmarks, early investigations indicate that large language models on their own do not improve on state-of-the-art results according to standard evaluation metrics. For grading it appears that linguistic features established in the literature should still be used for best performance, and for error correction it may be that the models can offer alternative feedback styles which are not measured sensitively with existing methods. In all cases, there is work to be done to experiment with the inclusion of large language models in education technology for language learners, in order to properly understand and report on their capacities and limitations, and to ensure that foreseeable risks such as misinformation and harmful bias are mitigated.
Large language models (LLMs) have demonstrated exceptional performance across a wide range of tasks. These models, powered by advanced deep learning techniques, have revolutionized the field of natural language processing (NLP) and have achieved remarkable results in various language-related tasks. LLMs have excelled in tasks such as machine translation, sentiment analysis, question answering, text generation, text classification, language modeling, and more. They have proven to be highly effective in capturing complex linguistic patterns, understanding context, and generating coherent and contextually relevant text. The attention scheme plays a crucial role in the architecture of large language models (LLMs). It is a fundamental component that enables the model to capture and utilize contextual information during language processing tasks effectively. Making the attention scheme computation faster is one of the central questions to speed up the LLMs computation. It is well-known that quantum machine has certain computational advantages compared to the classical machine. However, it is currently unknown whether quantum computing can aid in LLM. In this work, we focus on utilizing Grover's Search algorithm to compute a sparse attention computation matrix efficiently. We achieve a polynomial quantum speed-up over the classical method. Moreover, the attention matrix outputted by our quantum algorithm exhibits an extra low-rank structure that will be useful in obtaining a faster training algorithm for LLMs. Additionally, we present a detailed analysis of the algorithm's error analysis and time complexity within the context of computing the attention matrix.
Identifying implicit attitudes toward food can mitigate social prejudice due to food's salience as a marker of ethnic identity. Stereotypes about food are representational harms that may contribute to racialized discourse and negatively impact economic outcomes for restaurants. Understanding the presence of representational harms in online corpora in particular is important, given the increasing use of large language models (LLMs) for text generation and their tendency to reproduce attitudes in their training data. Through careful linguistic analyses, we evaluate social theories about attitudes toward immigrant cuisine in a large-scale study of framing differences in 2.1M English language Yelp reviews. Controlling for factors such as restaurant price and neighborhood racial diversity, we find that immigrant cuisines are more likely to be othered using socially constructed frames of authenticity (e.g., "authentic," "traditional"), and that non-European cuisines (e.g., Indian, Mexican) in particular are described as more exotic compared to European ones (e.g., French). We also find that non-European cuisines are more likely to be described as cheap and dirty, even after controlling for price, and even among the most expensive restaurants. Finally, we show that reviews generated by LLMs reproduce similar framing tendencies, pointing to the downstream retention of these representational harms. Our results corroborate social theories of gastronomic stereotyping, revealing racialized evaluative processes and linguistic strategies through which they manifest.
In healthcare, the ability to care for oneself is reflected in the "Activities of Daily Living (ADL)," which serve as a measure of functional ability (functioning). A lack of functioning may lead to poor living conditions requiring personal care and assistance. To accurately identify those in need of support, assistance programs continuously evaluate participants' functioning across various domains. However, the assessment process may encounter consistency issues when multiple assessors with varying levels of expertise are involved. Novice assessors, in particular, may lack the necessary preparation for real-world interactions with participants. To address this issue, we developed a dialogue system that simulates interactions between assessors and individuals of varying functioning in a natural and reproducible way. The dialogue system consists of two major modules, one for natural language understanding (NLU) and one for natural language generation (NLG), respectively. In order to generate responses consistent with the underlying knowledge base, the dialogue system requires both an understanding of the user's query and of biographical details of an individual being simulated. To fulfill this requirement, we experimented with query classification and generated responses based on those biographical details using some recently released InstructGPT-like models.
Text generation under constraints have seen increasing interests in natural language processing, especially with the rapidly improving capabilities of large language models. However, existing benchmarks for constrained generation usually focus on fixed constraint types (e.g.,generate a sentence containing certain words) that have proved to be easy for state-of-the-art models like GPT-4. We present COLLIE, a grammar-based framework that allows the specification of rich, compositional constraints with diverse generation levels (word, sentence, paragraph, passage) and modeling challenges (e.g.,language understanding, logical reasoning, counting, semantic planning). We also develop tools for automatic extraction of task instances given a constraint structure and a raw text corpus. Using COLLIE, we compile the COLLIE-v1 dataset with 2080 instances comprising 13 constraint structures. We perform systematic experiments across five state-of-the-art instruction-tuned language models and analyze their performances to reveal shortcomings. COLLIE is designed to be extensible and lightweight, and we hope the community finds it useful to develop more complex constraints and evaluations in the future.
In any system that uses structured knowledge graph (KG) data as its underlying knowledge representation, KG-to-text generation is a useful tool for turning parts of the graph data into text that can be understood by humans. Recent work has shown that models that make use of pretraining on large amounts of text data can perform well on the KG-to-text task even with relatively small sets of training data on the specific graph-to-text task. In this paper, we build on this concept by using large language models to perform zero-shot generation based on nothing but the model's understanding of the triple structure from what it can read. We show that ChatGPT achieves near state-of-the-art performance on some measures of the WebNLG 2020 challenge, but falls behind on others. Additionally, we compare factual, counter-factual and fictional statements, and show that there is a significant connection between what the LLM already knows about the data it is parsing and the quality of the output text.
Prompting large language models (LLMs) for data augmentation has recently become a common practice in few-shot NLP tasks. In this paper, we propose Chain-of-Thought Attribute Manipulation (CoTAM), a novel approach that generates new data from existing examples by only tweaking in the user-provided, task-specific attribute, e.g., sentiment polarity or topic in movie reviews. Instead of conventional latent representation controlling, we leverage the chain-of-thought prompting to directly edit the text in three steps, (1) attribute decomposition, (2) manipulation proposal, and (3) sentence reconstruction. Extensive results on various tasks, such as text (pair) classification, aspect-based sentiment analysis, and conditional text generation, verify the superiority of CoTAM over other LLM-based augmentation methods with the same number of training examples for both fine-tuning and in-context learning. Remarkably, the 2D visualization of the augmented dataset using principal component analysis revealed a human-recognizable decision boundary that is likely hinted by the attribute manipulation, demonstrating the potential of our proposed approach.
The dominant text generation models compose the output by sequentially selecting words from a fixed vocabulary. In this paper, we formulate text generation as progressively copying text segments (e.g., words or phrases) from an existing text collection. We compute the contextualized representations of meaningful text segments and index them using efficient vector search toolkits. The task of text generation is then decomposed into a series of copy-and-paste operations: at each time step, we seek suitable text spans from the text collection rather than selecting from a standalone vocabulary. Experiments on the standard language modeling benchmark (WikiText-103) show that our approach achieves better generation quality according to both automatic and human evaluations. Besides, its inference efficiency is comparable to token-level autoregressive models thanks to the reduction of decoding steps. We also show that our approach allows for effective domain adaptation by simply switching to domain-specific text collection without extra training. Finally, we observe that our approach attains additional performance gains by simply scaling up to larger text collections, again without further training.\footnote{Our source codes are publicly available at \url{https://github.com/gmftbyGMFTBY/Copyisallyouneed}.}
Existing evaluation metrics for natural language generation (NLG) tasks face the challenges on generalization ability and interpretability. Specifically, most of the well-performed metrics are required to train on evaluation datasets of specific NLG tasks and evaluation dimensions, which may cause over-fitting to task-specific datasets. Furthermore, existing metrics only provide an evaluation score for each dimension without revealing the evidence to interpret how this score is obtained. To deal with these challenges, we propose a simple yet effective metric called DecompEval. This metric formulates NLG evaluation as an instruction-style question answering task and utilizes instruction-tuned pre-trained language models (PLMs) without training on evaluation datasets, aiming to enhance the generalization ability. To make the evaluation process more interpretable, we decompose our devised instruction-style question about the quality of generated texts into the subquestions that measure the quality of each sentence. The subquestions with their answers generated by PLMs are then recomposed as evidence to obtain the evaluation result. Experimental results show that DecompEval achieves state-of-the-art performance in untrained metrics for evaluating text summarization and dialogue generation, which also exhibits strong dimension-level / task-level generalization ability and interpretability.