Malta

The escalating debate on AI&#39;s capabilities warrants developing reliable metrics to assess machine &quot;intelligence.&quot; Recently, many anecdotal examples were used to suggest that newer Large Language Models (LLMs) like ChatGPT and GPT-4 exhibit Neural Theory-of-Mind (N-ToM); however, prior work reached conflicting conclusions regarding those abilities. We investigate the extent of LLMs&#39; N-ToM through an extensive evaluation of 6 tasks and find that while LLMs exhibit certain N-ToM abilities, this behavior is far from being robust. We further examine the factors impacting performance on N-ToM tasks and discover that LLMs struggle with adversarial examples, indicating reliance on shallow heuristics rather than robust ToM abilities. We caution against drawing conclusions from anecdotal examples, limited benchmark testing, and using human-designed psychological tests to evaluate models.

EACL 2024

Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models

spurious-correlations

theory-of-mind

llms

The escalating debate on AI's capabilities warrants developing reliable metrics to assess machine "intelligence." Recently, many anecdotal examples were used to suggest that newer Large Language Models (LLMs) like ChatGPT and GPT-4 exhibit Neural Theory-of-Mind (N-ToM); however, prior work reached conflicting conclusions regarding those abilities. We investigate the extent of LLMs' N-ToM through an extensive evaluation of 6 tasks and find that while LLMs exhibit certain N-ToM abilities, this behavior is far from being robust. We further examine the factors impacting performance on N-ToM tasks and discover that LLMs struggle with adversarial examples, indicating reliance on shallow heuristics rather than robust ToM abilities. We caution against drawing conclusions from anecdotal examples, limited benchmark testing, and using human-designed psychological tests to evaluate models.

technical paper

Welcome to the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL). Continuing its mission of expanding and involving the science community of all European countries, EACL has selected the Malta community for the 18th EACL. Considering the importance of physical interaction among researchers, the conference will be held at the Hotel Radisson Blu, St. Julians, in Malta, from 17 to 22 of March, 2024. The conference will also feature the possibility to participate virtually. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language. 
#### [Conference Handbook](https://drive.google.com/file/d/13Wn38q0yev6U3RTgnf4Tn5T9GuAuZ-xW/view?usp=sharing) *downloadable*

To access the EACL 2024 event page you need to register [**here**](https://acl.swoogo.com/EACL2024). 

Please register!

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

We propose attribute-aware multimodal entity linking, where the input consists of a mention described with a text paragraph and images, and the goal is to predict the corresponding target entity from a multimodal knowledge base (KB) where each entity is also accompanied by a text description, visual images, and a collection of attributes that present the meta-information of the entity in a structured format. 
To facilitate this research endeavor, we construct Ameli, encompassing a new multimodal entity linking benchmark dataset that contains 16,735 mentions described in text and associated with 30,472 images, and a multimodal knowledge base that covers 34,690 entities along with 177,873 entity images and 798,216 attributes. To establish baseline performance on Ameli, we experiment with several state-of-the-art architectures for multimodal entity linking and further propose a new approach that incorporates attributes of entities into disambiguation. Experimental results and extensive qualitative analysis demonstrate that extracting and understanding the attributes of mentions from their text descriptions and visual images play a vital role in multimodal entity linking. To the best of our knowledge, we are the first to integrate attributes in the multimodal entity linking task. The programs, model checkpoints, and the dataset are publicly available at https://github.com/VT-NLP/Ameli.

Ameli: Enhancing Multimodal Entity Linking with Fine-Grained Attributes

Few-shot named entity recognition (NER) detects named entities within text using only a few annotated examples. One promising line of research is to leverage natural language descriptions of each entity type: the common label PER might, for example, be verbalized as ''person entity.'' In an initial label interpretation learning phase, the model learns to interpret such verbalized descriptions of entity types. In a subsequent few-shot tagset extension phase, this model is then given a description of a previously unseen entity type (such as ''music album'') and optionally a few training examples to perform few-shot NER for this type. In this paper, we systematically explore the impact of a strong semantic prior to interpret verbalizations of new entity types by massively scaling up the number and granularity of entity types used for label interpretation learning. To this end, we leverage an entity linking benchmark to create a dataset with orders of magnitude of more distinct entity types and descriptions as currently used datasets. We find that this increased signal yields strong results in zero- and few-shot NER in in-domain, cross-domain, and even cross-lingual settings. Our findings indicate significant potential for improving few-shot NER through heuristical data-based optimization.

Large-Scale Label Interpretation Learning for Few-Shot Named Entity Recognition

Accurately understanding temporal relations between events is a critical building block of diverse tasks, such as temporal reading comprehension (TRC) and relation extraction (TRE). For example in TRC, we need to understand the temporal semantic differences between the following two questions that are lexically near-identical: “What finished right before the decision?” or “What finished right after the decision?”. To discern the two questions, existing solutions have relied on answer overlaps as a proxy label to contrast similar and dissimilar questions. However, we claim that answer overlap can lead to unreliable results, due to spurious overlaps of two dissimilar questions with coincidentally identical answers. To address the issue, we propose a novel approach that elicits proper reasoning behaviors through a module for predicting time spans of events. We introduce the Timeline Reasoning Network (TRN) operating in a two-step inductive reasoning process: In the first step model initially answers each question with semantic and syntactic information. The next step chains multiple questions on the same event to predict a timeline, which is then used to ground the answers. Results on the TORQUE and TB-dense, TRC, and TRE tasks respectively, demonstrate that TRN outperforms previous methods by effectively resolving the spurious overlaps using the predicted timeline.

Chaining Event Spans for Temporal Relation Grounding

Recent studies have shown that as Transformer-based language models become larger and are trained on very large amounts of data, the fit of their surprisal estimates to naturalistic human reading times degrades. The current work presents a series of analyses showing that word frequency is a key explanatory factor underlying these two trends. First, residual errors from four language model families on four corpora show that the inverse correlation between model size and fit to reading times is the strongest on the subset of least frequent words, which is driven by excessively accurate predictions of larger model variants. Additionally, training dynamics reveal that during later training steps, all model variants learn to predict rare words and that larger model variants do so more accurately, which explains the detrimental effect of both training data amount and model size on fit to reading times. Finally, a feature attribution analysis demonstrates that larger model variants are able to accurately predict rare words based on both an effectively longer context window size as well as stronger local associations compared to smaller model variants. Taken together, these results indicate that Transformer-based language models' surprisal estimates diverge from human-like expectations due to the superhumanly complex associations they learn for predicting rare words.

Frequency Explains the Inverse Correlation of Large Language Models' Size, Training Data Amount, and Surprisal's Fit to Reading Times

Identification of cognates across related languages is one of the primary problems in historical linguistics. Automated cognate identification is helpful for several downstream tasks including identifying sound correspondences, proto-language reconstruction, phylogenetic classification, etc. Previous state-of-the-art methods are mostly based on distributions of phonemes computed across multilingual wordlists and make little use of the cognacy labels that define links among cognate clusters. In this paper, we present a transformer-based architecture inspired by computational biology for the task of automated cognate detection. Beyond a certain amount of supervision, this method performs better than the existing methods, and shows steady improvement with further increase in supervision proving the efficacy of utilizing the labeled information. We also demonstrate that accepting multiple sequence alignments as input and having an end-to-end architecture with link prediction head saves much computation time while simultaneously yielding superior performance.

Automated Cognate Detection as a Supervised Link Prediction Task with Cognate Transformer

Recent studies of the emergent capabilities of transformer-based Natural Language Understanding (NLU) models have indicated that they have an understanding of lexical and compositional semantics. We provide evidence that suggests these claims should be taken with a grain of salt: we find that state-of-the-art Natural Language Inference (NLI) models are sensitive towards minor semantics preserving surface-form variations, which lead to sizable inconsistent model decisions during inference. Notably, this behaviour differs from valid and in-depth comprehension of compositional semantics, however does neither emerge when evaluating model accuracy on standard benchmarks nor when probing for syntactic, monotonic, and logically robust reasoning. We propose a novel framework to measure the extent of semantic sensitivity. To this end, we evaluate NLI models on adversarially generated examples containing minor semantics-preserving surface-form input noise. This is achieved using conditional text generation, with the explicit condition that the NLI model predicts the relationship between the original and adversarial inputs as a symmetric equivalence entailment. We systematically study the effects of the phenomenon across NLI models for $\text{\emph{in-}}$ and $\text{\emph{out-of-}}$ domain settings. Our experiments show that semantic sensitivity causes performance degradations of $12.92\%$ and $23.71\%$ average over $\text{\emph{in-}}$ and $\text{\emph{out-of-}}$ domain settings, respectively. We further perform ablation studies, analysing this phenomenon across models, datasets, and variations in inference and show that semantic sensitivity can lead to major inconsistency within model predictions.

Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI Models

The propagation of offensive content through social media channels has garnered attention of the research community. Multiple works have proposed various semantically related yet subtle distinct categories of offensive speech. In this work, we explore meta-learning approaches to leverage the diversity of offensive speech corpora to enhance their reliable and efficient detection. We propose a joint embedding architecture that incorporates the input's label and definition for classification via Prototypical Network. Our model achieves at least 75% of the maximal F1-score while using less than 10% of the available training data across 4 datasets. Our experimental findings also provide a case study of training strategies valuable to combat resource scarcity.

“Define Your Terms” : Enhancing Efficient Offensive Speech Classification with Definition

With the ever-growing use of social media to express opinions on the national and international stage, unsupervised methods of stance detection are increasingly important to handle the task without costly annotation of data. The current unsupervised state-of-the-art models are designed for specific network types, either homophilic or heterophilic, and they fail to generalize to both. In this paper, we first analyze the generalization ability of recent baselines to these two very different network types. Then, we conduct extensive experiments with a baseline model based on text embeddings propagated with a graph neural network that generalizes well to heterophilic and homophilic networks. We show that it outperforms, on average, other state-of-the-art methods across the two network types. Additionally, we show that combining textual and network information outperforms using text only, and that the language model size has only a limited impact on the model performance.

Unsupervised stance detection for social media discussions: A generic baseline

Current text classification approaches usually focus on the content to be classified. Contextual aspects (both linguistic and extra-linguistic) are usually neglected, even in tasks based on online discussions. Still in many cases the multi-party and multi-turn nature of the context from which these elements are selected can be fruitfully exploited. In this work, we propose a series of experiments on a large dataset for stance detection in English, in which we evaluate the contribution of different types of contextual information, i.e. linguistic, structural and temporal, by feeding them as natural language input into a transformer-based model. We also experiment with different amounts of training data and analyse the topology of local discussion networks in a privacy-compliant way. Results show that structural information can be highly beneficial to text classification but only under certain circumstances (e.g. depending on the amount of training data and on discussion chain complexity). Indeed, we show that contextual information on smaller datasets from other classification tasks does not yield significant improvements. Our framework, based on local discussion networks, allows the integration of structural information while minimising user profiling, thus preserving their privacy.

Putting Context in Context: the Impact of Discussion Structure on Text Classification

We explore how weak supervision on abundant unlabeled data can be leveraged to improve few-shot performance in aspect-based sentiment analysis (ABSA) tasks. We propose a pipeline approach to construct a noisy ABSA dataset, and we use it to adapt a pre-trained sequence-to-sequence model to the ABSA tasks. We test the resulting model on three widely used ABSA datasets, before and after fine-tuning. Our proposed method preserves the full fine-tuning performance while showing significant improvements (15.84 absolute F1) in the few-shot learning scenario for the harder tasks. In zero-shot (i.e., without fine-tuning), our method outperforms the previous state of the art on the aspect extraction sentiment classification (AESC) task and is, additionally, capable of performing the harder aspect sentiment triplet extraction (ASTE) task.

Premium content

Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models

Downloads

Next from EACL 2024

Ameli: Enhancing Multimodal Entity Linking with Fine-Grained Attributes

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES