The wild thing!

Statistical language models improve the performance of speech recognition systems by providing estimates of a priori probabilities of word sequences. The commonly used trigram language models obtain the conditional probability estimate of a word given the previous two words, from a large corpus of text. The text corpus is often a collection of several small diverse segments such as newspaper articles, or conversations on different topics. Knowledge of the current topic could be utilized to adapt the general trigram language models to match that topic closely. For example, an interpolation of the general language model with one built on the topic data could be used. The authors first discuss the adaptation of general trigram language models to a known topic using the minimum discrimination information (MDI) method. They then present results on the switchboard corpus which consists of telephone conversations on several topics.

Despite the superior performance of large language models to generate natural language texts, it is hard to generate texts with correct logic according to a given task, due to the difficulties for neural models to capture strict logic from free-form texts. In this paper, we propose a novel graph-based language model, Logical-GLM, to extract strict logic from free-form texts and then infuse into language models. Specifically, we first capture information from natural language instructions and construct logical probability graphs that generally describe domains. Next, we generate logical skeletons to guide language model training, infusing domain knowledge into language models. At last, we alternately optimize the searching policy of graphs and language models until convergence. The experimental results show that Logical-GLM is both effective and efficient compared with traditional language models, despite using smaller-scale training data and fewer parameters. Our approach can generate instructional texts with more correct logic owing to the internalized domain knowledge. Moreover, the search of logical graphs reflects the inner mechanism of the language models, which improves the interpretability of black-box models.

Language model pruning is an essential technology for speech applications running on resource-constrained devices, and many pruning algorithms have been developed for conventional word n-gram models. However, while exponential language models can give superior performance, there has been little work on the pruning of these models. In this paper, we propose several pruning algorithms for general exponential language models. We show that our best algorithm applied to an exponential n-gram model outperforms existing n-gram model pruning algorithms by up to 0.4% absolute in speech recognition word-error rate on Wall Street Journal and Broadcast News data sets. In addition, we show that Model M, an exponential class-based language model, retains its performance improvement over conventional word n-gram models when pruned to equal size, with gains of up to 2.5% absolute in word-error rate.

A fast and memory-efficient N-gram language model lookup method for large vocabulary continuous speech recognition

Since Computer Science students are used to applying both General Purpose Programming Languages (GPPLs) and Domain-Specific Languages (DSLs), Generative Artificial Intelligence based on Language Models (LMs) can help them on automatic tasks, allowing them to focus on more creative tasks and higher skills. However, the teaching and evaluation of technical tasks in Computer Science can be inefficient and prone to errors. Thus, the main objective of this article is to explore the performance of LMs compared to that of undergraduate Computer Science students in a specific case study: designing and implementing RESTful APIs DSLs. This research aims to determine if LMs can enhance the efficiency and accuracy of these processes. Our case study involved 39 students and 5 different LMs that must use the two DSLs we also designed for their task assignment. To evaluate performance, we applied uniform criteria to student and LMs-generated solutions, enabling a comparative analysis of accuracy and effectiveness. With a case study comparing performance between students and LMs, this article contributes to checking to what extent LMs are able to carry out software development tasks involving the use of new DSLs specially designed for highly specific settings in a similar way as well-qualified Computer Science students are able to. The results underscore the importance of welldefined DSLs and effective prompting processes for optimal LM performance. Specifically, LMs demonstrated high variability in task execution, with two GPT-based LMs achieving similar grades to those scored by the best of the students for every task, obtaining 0.78 and 0.92 on a normalized scale [0, 1], with 0.23 and 0.14 Standard Deviation for ChatGPT-4 and ChatGPT-4o respectively. After the experience, we can conclude that a well-defined DSL and a proper prompting process, providing the LM with metadata, persistent prompts, and a good knowledge base, are crucial for good LM performance. When LMs receive the right prompts, both large and small LMs can achieve excellent results depending on the task.

The public release of ChatGPT, a generative artificial intelligence language model, caused wide-spread public interest in its abilities but also concern about the implications of the application on academia, depending on whether it was deemed benevolent (e.g., supporting analysis and simplification of tasks) or malevolent (e.g., assignment writing and academic misconduct). While ChatGPT has been shown to provide answers of sufficient quality to pass some university exams, its capacity to write essays that require an exploration of value concepts is unknown. This paper presents the results of a study where ChatGPT-4 (released May 2023) was tasked with writing a 1500-word essay to discuss the nature of values used in the assessment of cultural heritage significance. Based on an analysis of 36 iterations, ChatGPT wrote essays of limited length with about 50% of the stipulated word count being primarily descriptive and without any depth or complexity. The concepts, which are often flawed and suffer from inverted logic, are presented in an arbitrary sequence with limited coherence and without any defined line of argument. Given that it is a generative language model, ChatGPT often splits concepts and uses one or more words to develop tangential arguments. While ChatGPT provides references as tasked, many are fictitious, albeit with plausible authors and titles. At present, ChatGPT has the ability to critique its own work but seems unable to incorporate that critique in a meaningful way to improve a previous draft. Setting aside conceptual flaws such as inverted logic, several of the essays could possibly pass as a junior high school assignment but fall short of what would be expected in senior school, let alone at a college or university level.

We introduce a novel approach for building language models based on a systematic, recursive exploration of skip n-gram models which are interpolated using modified Kneser-Ney smoothing. Our approach generalizes language models as it contains the classical interpolation with lower order models as a special case. In this paper we motivate, formalize and present our approach. In an extensive empirical experiment over English text corpora we demonstrate that our generalized language models lead to a substantial reduction of perplexity between 3.1% and 12.7% in comparison to traditional language models using modified Kneser-Ney smoothing. Furthermore, we investigate the behaviour over three other languages and a domain specific corpus where we observed consistent improvements. Finally, we also show that the strength of our approach lies in its ability to cope in particular with sparse training data. Using a very small training data set of only 736 KB text we yield improvements of even 25.7% reduction of perplexity.

The advent of powerful transformer-based discriminative language models and, more recently, generative GPT-family models, has led to notable advancements in natural language processing (NLP), particularly in commonsense reasoning tasks. One such task is commonsense reasoning, where performance is usually evaluated through multiple-choice question-answering benchmarks. Till date, many such benchmarks have been proposed and `leaderboards' tracking state-of-the-art performance on those benchmarks suggest that transformer-based models are approaching human-like performance. However, due to documented problems such as hallucination and bias, the research focus is shifting from merely quantifying accuracy on the task to an in-depth, context-sensitive probing of LLMs' generalization and robustness. To gain deeper insight into diagnosing these models' performance in commonsense reasoning scenarios, this thesis addresses three main studies: the generalization ability of transformer-based language models on commonsense reasoning, the trend in confidence distribution of these language models confronted with ambiguous inference tasks, and a proposed risk-centric evaluation framework for both discriminative and generative language models.

In this paper we investigate whether a combination of topic specific language models can outperform a general purpose language model, using a trigram model as our baseline model. We show that in the ideal case -- in which it is known beforehand which model to use -- specific models perform considerably better than the baseline model. We test two methods that combine specific models and show that these combinations outperform the general purpose model, in particular if the data is diverse in terms of topics and vocabulary. Inspired by these findings, we propose to combine a decision tree and a set of dynamic Bayesian networks into a new model. The new model uses context information to dynamically select an appropriate specific model.

In this article, we propose a new language model processor named the C-Transformer to address the external memory bottleneck of language models. It consists of three key functional blocks: 1) homogeneous deep-neural-network (DNN)–Transformer/spiking-Transformer core (HDSC) with a hybrid multiplication/accumulation unit (HMAU) to enhance hardware utilization; 2) output spike speculation unit (OSSU) to increase the energy efficiency of spike-domain processing; and 3) implicit weight generation unit (IWGU) with extended sign compression (ESC) to eliminate the external memory bottleneck. The chip is fabricated in Samsung’s 28-nm 1P8M CMOS technology and operates at a supply voltage of 0.7–1.1 V with a maximum frequency of 200 MHz and supports various tasks such as language modeling, translation, and summarization. For models like generative pre-trained transformer 2 (GPT-2), multilingual text-to-text transfer transformer (mT5), text-to-text transfer transformer (T5), and FairSeq MachineTranslation (FSMT), the C-Transformer achieves 0.21×–0.33× computation energy and 0.37×–0.41× external memory access (EMA) energy compared to the baseline. Our chip demonstrates 13.6% lower energy consumption than the previous state-of-the-art, despite having 2.1× more parameters. Moreover, it consumes 63.8% less energy with a similar parameter size. The C-Transformer can complete various language model tasks with <1 s latency, notably FSMT in 0.09 s and GPT-2 in 0.656 s. By combining DNN–Transformer and Spiking-Transformer architectures, the C-Transformer enhances computational energy efficiency, eliminates external memory bottlenecks, and enables language models such as GPT-2 to achieve state-of-the-art performance on mobile devices.

Distinguishing hate speech from non-hate offensive language is challenging, as hate speech not always includes offensive slurs and offensive language not always express hate. Here, four deep learners based on the Bidirectional Encoder Representations from Transformers (BERT), with either general or domain-specific language models, were tested against two datasets containing tweets labelled as either ‘Hateful’, ‘Normal’ or ‘Offensive’. The results indicate that the attention-based models profoundly confuse hate speech with offensive and normal language. However, the pre-trained models outperform state-of-the-art results in terms of accurately predicting the hateful instances.

Introduction. Modern education faces a contradiction between the active integration of generative artificial intelligence and its underexplored potential for providing evaluative feedback in development students’ mathematical literacy. The purpose of the article is to identify the potential of using a generative language model as a teacher’s tool for generating expert-level evaluative feedback when assessing open-ended mathematical problems Materials and Methods. The research is based on systemic-activity, criteria-oriented, and comparative approaches. Methods employed included theoretical analysis of scholarly literature, criteria-based assessment combined with prompt engineering techniques, as well as quantitative and qualitative analysis to determine the agreement between the evaluative feedback generated by the language model and that provided by a human expert. The sample consisted of 51 students Results. The research experimentally confirmed the feasibility of using generative artificial intelligence for providing evaluative feedback in mathematics education. An effective strategy for automating the assessment of open-ended mathematical problems was developed and substantiated, based on criteria-based assessment and prompt engineering techniques using GigaChat Pro language model. Empirical data revealed a moderate agreement between the evaluative feedback generated by GigaChat Pro and that provided by an expert teacher: accuracy reached 73%, Cohen’s coefficient (k) was 0,57, and the semantic similarity of textual comments (BertScore F1) was 0,614.

Large language models have demonstrated remarkable capabilities in facilitating general human-computer interactions. However, their effectiveness in providing domain-specific information in professional fields is limited. In this paper, we present an intelligent assisted decision-making framework utilizing large-language models to provide professional advice in taxation. By augmenting the large language model with a domain-specific database, we aim to overcome the limitations of generic language models and provide professional advice for tax-related queries. The results of our study indicate that the proposed framework sur-passes the credibility of generic large language models by providing domain-specific advice in taxation.

Designing intelligent systems to classify text in the medical domain is a challenging task. There is a shortage of openly available medical datasets (due to HIPPA-related strict regulations on protected health information for patients). In this paper, we explore the application of Open Source Medical LLMs (such as Meditron LLM), generic Large Language Models (such as LLAMA2), and Small Language Models (such as Phi2) on medical text classification (medical abstract dataset). We show that PEFT approaches such as LoRA can perform very well in classifying medical text, which involves interpreting patient conditions and symptoms and determining what medical problems the patients have. These approaches (based on Large and Small Language Models) have outperformed the current state of the results on medical abstracts corpus. In addition to medical LLMs, the open-source generic LLMs can be adapted to solving classification tasks on medical text and perform nearly as well as the specialized medical LLMs. SLMs can be a serious contender for solving domain-specific classification tasks (e.g., medical literature). This shows that carefully selecting the training data and fine-tuning positively impacts classification accuracy, precision, and recall. Generic Language Models such as LLAMA2 (LLM) and Phi2 (SLM) weren’t specifically trained with medical text. Medical LLMs such as Meditron outperform LLAMA2 and Phi2 in precision and accuracy. This is quite evident as Meditron was originally trained on medical text. The (micro averaged) F1 score for the fine-tuned Meditron model is 0.64. This is superior compared to fined-tuned LLAMA2 of 0.58 and Phi2 of 0.62. We see that Phi2 can outperform LLAMA2 with fewer number of parameters. The approaches used in this work can be extended to other medical text classification problems in the medical domain.

Generative language models (LMs) offer numerous advantages but may produce inappropriate or harmful outputs due to the harmful knowledge acquired during pre-training. This knowledge often manifests as undesirable correspondences, such as “harmful prompts” leading to “harmful outputs,” which our research aims to mitigate through unlearning techniques.However, existing unlearning methods based on gradient ascent can significantly impair the performance of LMs. To address this issue, we propose a novel approach called Weighted Positional N-pair (WPN) Learning, which leverages position-weighted mean pooling within an n-pair contrastive learning framework. WPN is designed to modify the output distribution of LMs by eliminating specific harmful outputs (e.g., replacing toxic responses with neutral ones), thereby transforming the model’s behavior from “harmful prompt-harmful output” to “harmful prompt-harmless response”. Experiments on OPT and GPT-NEO LMs show that WPN effectively reduces the proportion of harmful responses, achieving a harmless rate of up to 95.8% while maintaining stable performance on nine common benchmarks (with less than 2% degradation on average). Moreover, we provide empirical evidence to demonstrate WPN’s ability to weaken the harmful correspondences in terms of generalizability and robustness, as evaluated on out-of-distribution test sets and under adversarial attacks.