Achieving precise semantic control over the latent spaces of Variational AutoEncoders (VAEs) holds significant value for downstream tasks in NLP as the underlying generative mechanisms could be better localised, explained and improved upon. Recent research, however, has struggled to achieve consistent results, primarily due to the inevitable loss of semantic information in the variational bottleneck and limited control over the decoding mechanism. To overcome these challenges, we investigate discrete latent spaces in Vector Quantized Variational AutoEncoder (VQVAE) to improve semantic control and generation in Transformer-based VAEs. In particular, We propose T5VQVAE, a novel model that leverages the controllability of VQVAE to guide the self-attention mechanism in T5, exploiting its full generalization capabilities. Experimental results indicate that T5VQVAE outperforms existing state-of-the-art VAE models, including Optimus, in terms of control and preservation of semantic information across different tasks such as auto-encoding of sentences and mathematical expressions, text transfer, and inference. Moreover, T5VQVAE exhibits improved reasoning capabilities, suggesting potential applications for downstream natural language and symbolic inference tasks.

The injection of syntactic information in Variational AutoEncoders (VAEs) can result in an overall improvement of performances and generalisation. An effective strategy to achieve such a goal is to separate the encoding of distributional semantic features and syntactic structures into heterogeneous latent spaces via multi-task learning or dual encoder architectures. However, existing works employing such techniques are limited to LSTM-based VAEs. This work investigates latent space separation methods for structural syntactic injection in Transformer-based VAE architectures (i.e., Optimus) through the integration of graph-based models. Our empirical evaluation reveals that the proposed end-to-end VAE architecture can improve theoverall organisation of the latent space, alleviating the information loss occurring in standard VAE setups, and resulting in enhanced performances on language modelling and downstream generation tasks.

Efforts to apply transformer-based language models (TLMs) to the problem of reasoning in natural language have enjoyed ever-increasing success in recent years. The most fundamental task in this area to which nearly all others can be reduced is that of determining satisfiability. However, from a logical point of view, satisfiability problems vary along various dimensions, which may affect TLMs’ ability to learn how to solve them. The problem instances of satisfiability in natural language can belong to different computational complexity classes depending on the language fragment in which they are expressed. Although prior research has explored the problem of natural language satisfiability, the above-mentioned point has not been discussed adequately. Hence, we investigate how problem instances from varying computational complexity classes and having different grammatical constructs impact TLMs’ ability to learn rules of inference. Furthermore, to faithfully evaluate TLMs, we conduct an empirical study to explore the distribution of satisfiability problems.

How do different generalised quantifiers affect the behaviour of transformer-based language models (TLMs)? The recent popularity of TLMs and the central role generalised quantifiers have traditionally played in linguistics and logic bring this question into particular focus. The current research investigating this subject has not utilised a task defined purely in a logical sense, and thus, has not captured the underlying logical significance of generalised quantifiers. Consequently, they have not answered the aforementioned question faithfully or adequately. Therefore, we investigate how different generalised quantifiers affect TLMs by employing a textual entailment problem defined in a purely logical sense, namely, model-checking with natural language. Our approach permits the automatic construction of datasets with respect to which we can assess the ability of TLMs to learn the meanings of generalised quantifiers. Our investigation reveals that TLMs generally can comprehend the logical semantics of the most common generalised quantifiers, but that distinct quantifiers influence TLMs in varying ways.

Can transformers learn to comprehend logical semantics in natural language? Although many strands of work on natural language inference have focussed on transformer models’ ability to perform reasoning on text, the above question has not been answered adequately. This is primarily because the logical problems that have been studied in the context of natural language inference have their computational complexity vary with the logical and grammatical constructs within the sentences. As such, it is difficult to access whether the difference in accuracy is due to logical semantics or the difference in computational complexity. A problem that is much suited to address this issue is that of the model-checking problem, whose computational complexity remains constant (for fragments derived from first-order logic). However, the model-checking problem remains untouched in natural language inference research. Thus, we investigated the problem of model-checking with natural language to adequately answer the question of how the logical semantics of natural language affects transformers’ performance. Our results imply that the language fragment has a significant impact on the performance of transformer models. Furthermore, we hypothesise that a transformer model can at least partially understand the logical semantics in natural language but can not completely learn the rules governing the model-checking algorithm.

State-of-the-art deep-learning-based approaches to Natural Language Processing (NLP) are credited with various capabilities that involve reasoning with natural language texts. %However, reasoning in this setting is often ill-defined and shallow. In this paper we carry out a large-scale empirical study investigating the detection of formally valid inferences in controlled fragments of natural language for which the satisfiability problem becomes increasingly complex. We find that, while transformer-based language models perform surprisingly well in these scenarios, a deeper analysis reveals that they appear to overfit to superficial patterns in the data rather than acquiring the logical principles governing the reasoning in these fragments.

An emerging line of research in Explainable NLP is the creation of datasets enriched with human-annotated explanations and rationales, used to build and evaluate models with step-wise inference and explanation generation capabilities. While human-annotated explanations are used as ground-truth for the inference, there is a lack of systematic assessment of their consistency and rigour. In an attempt to provide a critical quality assessment of Explanation Gold Standards (XGSs) for NLI, we propose a systematic annotation methodology, named Explanation Entailment Verification (EEV), to quantify the logical validity of human-annotated explanations. The application of EEV on three mainstream datasets reveals the surprising conclusion that a majority of the explanations, while appearing coherent on the surface, represent logically invalid arguments, ranging from being incomplete to containing clearly identifiable logical errors. This conclusion confirms that the inferential properties of explanations are still poorly formalised and understood, and that additional work on this line of research is necessary to improve the way Explanation Gold Standards are constructed.

The relational syllogistic is an extension of the language of Classical syllogisms in which predicates are allowed to feature transitive verbs with quantified objects. It is known that the relational syllogistic does not admit a finite set of syllogism-like rules whose associated (direct) derivation relation is sound and complete. We present a modest extension of this language which does.