Human speakers can generate descriptions of perceptual concepts, abstracted from the instance-level. Moreover, such descriptions can be used by other speakers to learn provisional representations of those concepts. Learning and using abstract perceptual concepts is under-investigated in the language-and-vision field. The problem is also highly relevant to the field of representation learning in multi-modal NLP. In this paper, we introduce a framework for testing category-level perceptual grounding in multi-modal language models. In particular, we train separate neural networks to **generate** and **interpret** descriptions of visual categories. We measure the *communicative success* of the two models with the zero-shot classification performance of the interpretation model, which we argue is an indicator of perceptual grounding. Using this framework, we compare the performance of *prototype*- and *exemplar*-based representations. Finally, we show that communicative success exposes performance issues in the generation model, not captured by traditional intrinsic NLG evaluation metrics, and argue that these issues stem from a failure to properly ground language in vision at the category level.

We propose the VDG Challenge: a shared task that addresses and benchmarks the task of utterance generation in collaborative visual dialogue. The task features two challenging datasets, an evaluation protocol and a tentative schedule. Our shared task will allow researchers to unravel problems of modelling multi-modal interaction and fit of the existing approaches in the NLP and NLG communities.

Internal representations in transformer models can encode useful linguistic knowledge about syntax. Such knowledge could help optimise the data annotation process. However, identifying and extracting such representations from big language models is challenging. In this paper we evaluate two multilingual transformers for the presence of knowledge about the syntactic complexity of sentences and examine different vector norms. We provide a fine-grained evaluation of different norms in different layers and for different languages. Our results suggest that no single part in the models would be the primary source for the knowledge of syntactic complexity. But some norms show a higher degree of sensitivity to syntactic complexity, depending on the language and model used.

Language-and-vision models have shown good performance in tasks such as image-caption matching and caption generation. However, it is challenging for such models to generate pragmatically correct captions, which adequately reflect what is happening in one image or several images. It is crucial to evaluate this behaviour to understand underlying reasons behind it. Here we explore to what extent contextual language-and-vision models are sensitive to different discourse, both textual and visual. In particular, we employ one of the multi-modal transformers (ViLBERT) and test if it can match descriptions and images, differentiating them from distractors of different degree of similarity that are sampled from different visual and textual contexts. We place our evaluation in the multi-sentence and multi-image setup, where images and sentences are expected to form a single narrative structure. We show that the model can distinguish different situations but it is not sensitive to differences within one narrative structure. We also show that performance depends on the task itself, for example, what modality remains unchanged in non-matching pairs or how similar non-matching pairs are to original pairs.

This paper describes insights into how different inference algorithms structure discourse in image paragraphs. We train a multi-modal transformer and compare 11 variations of decoding algorithms. We propose to evaluate image paragraphs not only with standard automatic metrics, but also with a more extensive, “under the hood” analysis of the discourse formed by sentences. Our results show that while decoding algorithms can be unfaithful to the reference texts, they still generate grounded descriptions, but they also lack understanding of the discourse structure and differ from humans in terms of attentional structure over images.

We explore how a multi-modal transformer trained for generation of longer image descriptions learns syntactic and semantic representations about entities and relations grounded in objects at the level of masked self-attention (text generation) and cross-modal attention (information fusion). We observe that cross-attention learns the visual grounding of noun phrases into objects and high-level semantic information about spatial relations, while text-to-text attention captures low-level syntactic knowledge between words. This concludes that language models in a multi-modal task learn different semantic information about objects and relations cross-modally and uni-modally (text-only). Our code is available here: https://github.com/GU-CLASP/attention-as-grounding.

In this paper we examine different meaning representations that are commonly used in different natural language applications today and discuss their limits, both in terms of the aspects of the natural language meaning they are modelling and in terms of the aspects of the application for which they are used.

We investigate how different augmentation techniques on both textual and visual representations affect the performance of the face description generation model. Specifically, we provide the model with either original images, sketches of faces, facial composites or distorted images. In addition, on the language side, we experiment with different methods to augment the original dataset with paraphrased captions, which are semantically equivalent to the original ones, but differ in terms of their form. We also examine if augmenting the dataset with descriptions from a different domain (e.g., image captions of real-world images) has an effect on the performance of the models. We train models on different combinations of visual and linguistic features and perform both (i) automatic evaluation of generated captions and (ii) examination of how useful different visual features are for the task of facial feature classification. Our results show that although original images encode the best possible representation for the task, the model trained on sketches can still perform relatively well. We also observe that augmenting the dataset with descriptions from a different domain can boost performance of the model. We conclude that face description generation systems are more susceptible to language rather than vision data augmentation. Overall, we demonstrate that face caption generation models display a strong imbalance in the utilisation of language and vision modalities, indicating a lack of proper information fusion. We also describe ethical implications of our study and argue that future work on human face description generation should create better, more representative datasets.

The problem of interpretation of knowledge learned by multi-head self-attention in transformers has been one of the central questions in NLP. However, a lot of work mainly focused on models trained for uni-modal tasks, e.g. machine translation. In this paper, we examine masked self-attention in a multi-modal transformer trained for the task of image captioning. In particular, we test whether the multi-modality of the task objective affects the learned attention patterns. Our visualisations of masked self-attention demonstrate that (i) it can learn general linguistic knowledge of the textual input, and (ii) its attention patterns incorporate artefacts from visual modality even though it has never accessed it directly. We compare our transformer’s attention patterns with masked attention in distilgpt-2 tested for uni-modal text generation of image captions. Based on the maps of extracted attention weights, we argue that masked self-attention in image captioning transformer seems to be enhanced with semantic knowledge from images, exemplifying joint language-and-vision information in its attention patterns.

The RDF-to-text task has recently gained substantial attention due to the continuous growth of RDF knowledge graphs in number and size. Recent studies have focused on systematically comparing RDF-to-text approaches on benchmarking datasets such as WebNLG. Although some evaluation tools have already been proposed for text generation, none of the existing solutions abides by the Findability, Accessibility, Interoperability, and Reusability (FAIR) principles and involves RDF data for the knowledge extraction task. In this paper, we present BENG, a FAIR benchmarking platform for Natural Language Generation (NLG) and Knowledge Extraction systems with focus on RDF data. BENG builds upon the successful benchmarking platform GERBIL, is opensource and is publicly available along with the data it contains.

WebNLG+ offers two challenges: (i) mapping sets of RDF triples to English or Russian text (generation) and (ii) converting English or Russian text to sets of RDF triples (semantic parsing). Compared to the eponymous WebNLG challenge, WebNLG+ provides an extended dataset that enable the training, evaluation, and comparison of microplanners and semantic parsers. In this paper, we present the results of the generation and semantic parsing task for both English and Russian and provide a brief description of the participating systems.

Generating multi-sentence image descriptions is a challenging task, which requires a good model to produce coherent and accurate paragraphs, describing salient objects in the image. We argue that multiple sources of information are beneficial when describing visual scenes with long sequences. These include (i) perceptual information and (ii) semantic (language) information about how to describe what is in the image. We also compare the effects of using two different pooling mechanisms on either a single modality or their combination. We demonstrate that the model which utilises both visual and language inputs can be used to generate accurate and diverse paragraphs when combined with a particular pooling mechanism. The results of our automatic and human evaluation show that learning to embed semantic information along with visual stimuli into the paragraph generation model is not trivial, raising a variety of proposals for future experiments.

We present a dataset consisting of what we call image description sequences, which are multi-sentence descriptions of the contents of an image. These descriptions were collected in a pseudo-interactive setting, where the describer was told to describe the given image to a listener who needs to identify the image within a set of images, and who successively asks for more information. As we show, this setup produced nicely structured data that, we think, will be useful for learning models capable of planning and realising such description discourses.

Image captioning models are typically trained on data that is collected from people who are asked to describe an image, without being given any further task context. As we argue here, this context independence is likely to cause problems for transferring to task settings in which image description is bound by task demands. We demonstrate that careful design of data collection is required to obtain image descriptions which are contextually bounded to a particular meta-level task. As a task, we use MeetUp!, a text-based communication game where two players have the goal of finding each other in a visual environment. To reach this goal, the players need to describe images representing their current location. We analyse a dataset from this domain and show that the nature of image descriptions found in MeetUp! is diverse, dynamic and rich with phenomena that are not present in descriptions obtained through a simple image captioning task, which we ran for comparison.