Open English Wordnet is a key resource published in OntoLex-lemon as part of the linguistic linked open data cloud. There are, however, many links missing in the resource, and in this paper, we look at how we can establish hypernymy between adjectives. We present a theoretical discussion of the hypernymy relation and how it differs for adjectives in contrast to nouns and verbs. We develop a new resource for adjective hypernymy and fine-tune large language models to predict adjective hypernymy, showing that the methodology of TaxoLLaMa can be adapted to this task.

We investigate the training dynamics of deep classifiers by examining how hierarchical relationships between classes evolve during training. Through extensive experiments, we argue that the learning process in classification problems can be understood through the lens of label clustering. Specifically, we observe that networks tend to distinguish higher-level (hypernym) categories in the early stages of training, and learn more specific (hyponym) categories later. We introduce a novel framework to track the evolution of the feature manifold during training, revealing how the hierarchy of class relations emerges and refines across the network layers. Our analysis demonstrates that the learned representations closely align with the semantic structure of the dataset, providing a quantitative description of the clustering process. Notably, we show that in the hypernym label space, certain properties of neural collapse appear earlier than in the hyponym label space, helping to bridge the gap between the initial and terminal phases of learning. We believe our findings offer new insights into the mechanisms driving hierarchical learning in deep networks, paving the way for future advancements in understanding deep learning dynamics.

Adapting named entity recognition (NER) methods to new domains poses significant challenges. We introduce RapidNER, a framework designed for the rapid deployment of NER systems through efficient dataset construction. RapidNER operates through three key steps: (1) extracting domain-specific sub-graphs and triples from a general knowledge graph, (2) collecting and leveraging texts from various sources to build the NERsocial dataset, which focuses on entities typical in human-robot interaction, and (3) implementing an annotation scheme using Elasticsearch (ES) to enhance efficiency. NERsocial, validated by human annotators, includes six entity types, 153K tokens, and 99.4K sentences, demonstrating RapidNER's capability to expedite dataset creation.

Verbs form the backbone of language, providing the structure and meaning to sentences. Yet, their intricate semantic nuances pose a longstanding challenge. Understanding verb relations through the concept of lexical entailment is crucial for comprehending sentence meanings and grasping verb dynamics. This work investigates the capabilities of eight Large Language Models in recognizing lexical entailment relations among verbs through differently devised prompting strategies and zero-/few-shot settings over verb pairs from two lexical databases, namely WordNet and HyperLex. Our findings unveil that the models can tackle the lexical entailment recognition task with moderately good performance, although at varying degree of effectiveness and under different conditions. Also, utilizing few-shot prompting can enhance the models' performance. However, perfectly solving the task arises as an unmet challenge for all examined LLMs, which raises an emergence for further research developments on this topic.

Prompting serves as the major way humans interact with Large Language Models (LLM). Commercial AI systems commonly define the role of the LLM in system prompts. For example, ChatGPT uses "You are a helpful assistant" as part of the default system prompt. But is "a helpful assistant" the best role for LLMs? In this study, we present a systematic evaluation of how social roles in system prompts affect model performance. We curate a list of 162 roles covering 6 types of interpersonal relationships and 8 types of occupations. Through extensive analysis of 3 popular LLMs and 2457 questions, we show that adding interpersonal roles in prompts consistently improves the models' performance over a range of questions. Moreover, while we find that using gender-neutral roles and specifying the role as the audience leads to better performances, predicting which role leads to the best performance remains a challenging task, and that frequency, similarity, and perplexity do not fully explain the effect of social roles on model performances. Our results can help inform the design of system prompts for AI systems. Code and data are available at https://github.com/Jiaxin-Pei/Prompting-with-Social-Roles.

Concepts play a pivotal role in various human cognitive functions, including learning, reasoning and communication. However, there is very little work on endowing machines with the ability to form and reason with concepts. In particular, state-of-the-art large language models (LLMs) work at the level of tokens, not concepts. In this work, we analyze how well contemporary LLMs capture human concepts and their structure. We then discuss ways to develop concept-aware LLMs, taking place at different stages of the pipeline. We sketch a method for pretraining LLMs using concepts, and also explore the simpler approach that uses the output of existing LLMs. Despite its simplicity, our proof-of-concept is shown to better match human intuition, as well as improve the robustness of predictions. These preliminary results underscore the promise of concept-aware LLMs.

Text-to-image synthesis has recently attracted widespread attention due to rapidly improving quality and numerous practical applications. However, the language understanding capabilities of text-to-image models are still poorly understood, which makes it difficult to reason about prompt formulations that a given model would understand well. In this work, we measure the capability of popular text-to-image models to understand $\textit{hypernymy}$, or the "is-a" relation between words. We design two automatic metrics based on the WordNet semantic hierarchy and existing image classifiers pretrained on ImageNet. These metrics both enable broad quantitative comparison of linguistic capabilities for text-to-image models and offer a way of finding fine-grained qualitative differences, such as words that are unknown to models and thus are difficult for them to draw. We comprehensively evaluate popular text-to-image models, including GLIDE, Latent Diffusion, and Stable Diffusion, showing how our metrics can provide a better understanding of the individual strengths and weaknesses of these models.

Taxonomy expansion task is essential in organizing the ever-increasing volume of new concepts into existing taxonomies. Most existing methods focus exclusively on using textual semantics, leading to an inability to generalize to unseen terms and the "Prototypical Hypernym Problem." In this paper, we propose Visual Taxonomy Expansion (VTE), introducing visual features into the taxonomy expansion task. We propose a textual hypernymy learning task and a visual prototype learning task to cluster textual and visual semantics. In addition to the tasks on respective modalities, we introduce a hyper-proto constraint that integrates textual and visual semantics to produce fine-grained visual semantics. Our method is evaluated on two datasets, where we obtain compelling results. Specifically, on the Chinese taxonomy dataset, our method significantly improves accuracy by 8.75 %. Additionally, our approach performs better than ChatGPT on the Chinese taxonomy dataset.

Pre-trained language models (PLMs) were considered to be able to store relational knowledge present in the training data. However, some relational knowledge seems to be discarded unsafely in PLMs due to \textbf{report bias}: low-frequency relational knowledge might be underexpressed compared to high-frequency one in PLMs. This gives us a hint that relational knowledge might not be redundant to the stored knowledge of PLMs, but rather be complementary. To additionally inject relational knowledge into PLMs, we propose a simple-yet-effective approach to inject relational knowledge into PLMs, which is inspired by three observations (namely, polymorphism, synonymous substitution, and association). In particular, we switch entities in the training corpus to related entities (either hypernyms/hyponyms/synonyms, or arbitrarily-related concepts). Experimental results show that the proposed approach could not only better capture relational knowledge, but also improve the performance in various biomedical downstream tasks. Our model is available in \url{https://github.com/StevenZHB/BioPLM_InjectingKnowledge}.

We examine LMs' competence of directional predicate entailments by supervised fine-tuning with prompts. Our analysis shows that contrary to their apparent success on standard NLI, LMs show limited ability to learn such directional inference; moreover, existing datasets fail to test directionality, and/or are infested by artefacts that can be learnt as proxy for entailments, yielding over-optimistic results. In response, we present BoOQA (Boolean Open QA), a robust multi-lingual evaluation benchmark for directional predicate entailments, extrinsic to existing training sets. On BoOQA, we establish baselines and show evidence of existing LM-prompting models being incompetent directional entailment learners, in contrast to entailment graphs, however limited by sparsity.