Malta

With the rise of increasingly powerful and user-facing NLP systems, there is growing interest in assessing whether they have a good _representation of uncertainty_ by evaluating the quality of their predictive distribution over outcomes. We identify two main perspectives that drive starkly different evaluation protocols. The first treats predictive probability as an indication of model confidence; the second as an indication of human label variation. We discuss their merits and limitations, and take the position that both are crucial for trustworthy and fair NLP systems, but that exploiting a single predictive distribution is limiting. We recommend tools and highlight exciting directions towards models with disentangled representations of uncertainty about predictions and uncertainty about human labels.

EACL 2024

Interpreting Predictive Probabilities: Model Confidence or Human Label Variation?

label variation

variability

uncertainty

confidence

poster

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.

Inductive reasoning is fundamental to both human and artificial intelligence. 
The inductive reasoning abilities of current Large Language Models (LLMs) are evaluated in this research.
We argue that only considering induction of rules is too narrow and unrealistic, since inductive reasoning is usually mixed with other abilities, like rules application, results/rules validation, and updated information integration.
We probed the LLMs with a set of designed symbolic tasks and found that even state-of-the-art (SotA) LLMs fail significantly, showing the inability of LLMs to perform these intuitively simple tasks.
Furthermore, we found that perfect accuracy in a small-size problem does not guarantee the same accuracy in a larger-size version of the same problem, provoking the question of how we can assess the LLMs' actual problem-solving capabilities.
We also argue that Chain-of-Thought prompts help the LLMs by decomposing the problem-solving process, but the LLMs still learn limitedly.
Furthermore, we reveal that few-shot examples assist LLM generalization in out-of-domain (OOD) cases, albeit limited. The LLM starts to fail when the problem deviates from the provided few-shot examples.

A Comprehensive Evaluation of Inductive Reasoning Capabilities and Problem Solving in Large Language Models

Neural Machine Translation (NMT) models have been shown to be vulnerable to adversarial attacks, wherein carefully crafted perturbations of the input can mislead the target model. In this paper, we introduce ACT, a novel adversarial attack framework against NMT systems guided by a classifier. In our attack, the adversary aims to craft meaning-preserving adversarial examples whose translations in the target language by the NMT model belong to a different class than the original translations. Unlike previous attacks, our new approach has a more substantial effect on the translation by altering the overall meaning, which then leads to a different class determined by an oracle classifier. To evaluate the robustness of NMT models to our attack, we propose enhancements to existing black-box word-replacement-based attacks by incorporating output translations of the target NMT model and the output logits of a classifier within the attack process. Extensive experiments, including a comparison with existing untargeted attacks, show that our attack is considerably more successful in altering the class of the output translation and has more effect on the translation. This new paradigm can reveal the vulnerabilities of NMT systems by focusing on the class of translation rather than the mere translation quality as studied traditionally.

A Classification-Guided Approach for Adversarial Attacks against Neural Machine Translation

Sequence-to-sequence models often require an expensive autoregressive decoding process. However, for some downstream tasks such as out-of-distribution (OOD) detection and resource allocation, the actual decoding output is not needed, just a scalar attribute of this sequence. In such scenarios, where knowing the quality of a system's output to predict poor performance prevails over knowing the output itself, is it possible to bypass the autoregressive decoding? We propose Non-Autoregressive Proxy (NAP) models that can efficiently predict scalar-valued sequence-level attributes. Importantly, NAPs predict these metrics directly from the encodings, avoiding the expensive decoding stage. We consider two sequence tasks: Machine Translation (MT) and Automatic Speech Recognition (ASR). In OOD for MT, NAPs outperform ensembles while being significantly faster. NAPs are also proven capable of predicting metrics such as BERTScore (MT) or word error rate (ASR). For downstream tasks, such as data filtering and resource optimization, NAPs generate performance predictions that outperform predictive uncertainty while being highly inference efficient.

Who Needs Decoders? Efficient Estimation of Sequence-Level Attributes with Proxies

The knowledge encapsulated in a model is the core factor determining its final performance on downstream tasks. 
Much research in NLP has focused on efficient methods for storing and adapting different types of knowledge, e.g., in dedicated modularized structures, and on how to effectively combine these, e.g., by learning additional parameters. However, given the many possible options, a thorough understanding of the mechanisms involved in these compositions is missing, and hence it remains unclear which strategies to utilize. To address this research gap, we propose a novel framework for zero-shot module composition, which encompasses existing and some novel variations for selecting, weighting, and combining parameter modules under a single unified notion. Focusing on the scenario of domain knowledge and adapter layers, our framework provides a systematic unification of concepts, allowing us to conduct the first comprehensive benchmarking study of various zero-shot knowledge composition strategies. In particular, we test two module combination methods and five selection and weighting strategies for their effectiveness and efficiency in an extensive experimental setup. Our results highlight the efficacy of ensembling but also hint at the power of simple though often-ignored weighting methods. Further in-depth analyses allow us to understand the role of weighting vs. top-k selection, and show that, to a certain extent, the performance of adapter composition can even be predicted.

What the Weight?! A Unified Framework for Zero-Shot Knowledge Composition

Keyphrase extraction is the task of identifying a set of keyphrases present in a document that captures its most salient topics. Scientific domain-specific pre-training has led to achieving state-of-the-art keyphrase extraction performance with a majority of benchmarks being within the domain. In this work, we explore how to effectively enable the cross-domain generalization capabilities of such models without requiring the same scale of data. We primarily focus on the few-shot setting in non-scientific domain datasets such as OpenKP from the Web domain & StackEx from the StackExchange forum. We propose to leverage topic information intrinsically available in the data, to build a novel clustering-based sampling approach that facilitates selecting a few samples to label from the target domain facilitating building robust and performant models. This approach leads to large gains in performance of up to 26.35 points in F1 when compared to selecting few-shot samples uniformly at random. We also explore the setting where we have access to labeled data from the model's pretraining domain corpora and perform gradual training which involves slowly folding in target domain data to the source domain data. Here we demonstrate further improvements in the model performance by up to 12.76 F1 points.

Clustering-based Sampling for Few-Shot Cross-Domain Keyphrase Extraction

Siamese encoders such as sentence transformers are among the least understood deep models.
Established attribution methods cannot tackle this model class since it compares two inputs rather than processing a single one. 
To address this gap, we have recently proposed an attribution method specifically for Siamese encoders (Möller et al., 2023). 
However, it requires models to be adjusted and fine-tuned and therefore cannot be directly applied to off-the-shelf models. 
In this work, we reassess these restrictions and propose (i) a model with exact attribution ability that retains the original model’s predictive performance and (ii) a way to compute approximate attributions for off-the-shelf models.
We extensively compare approximate and exact attributions and use them to analyze the models’ attendance to different linguistic aspects. We gain insights into which syntactic roles Siamese transformers attend to, confirm that they mostly ignore negation, explore how they judge semantically opposite adjectives, and find that they exhibit lexical bias.

Approximate Attributions for Off-the-Shelf Siamese Transformers

Large Language Models (LLMs) may hallucinate and generate fake information, despite pre-training on factual data. Inspired by the journalistic device of ``according to sources'', we propose according-to prompting: directing LLMs to ground responses against previously observed text. To quantify this grounding, we propose a novel evaluation metric (QUIP-Score) that measures the extent to which model-produced answers are directly found in underlying text corpora. We illustrate with experiments on three corpora (Wikipedia, PubMed, and the U.S. legal tax code) that these prompts improve grounding under our metrics, with the additional benefit of often improving end-task performance. Furthermore, prompts that ask the model to decrease grounding (or to ground to other corpora) indeed {decrease} QUIP-Score, indicating the ability of LLMs to increase or decrease grounded generations on request.

“According to . . . ”: Prompting Language Models Improves Quoting from Pre-Training Data

Large language models (LLMs) have revolutionized natural language processing (NLP) by excelling at understanding and generating human-like text. However, their widespread deployment can be prohibitively expensive. SortedNet is a recent training technique for enabling dynamic inference by leveraging the modularity in networks and sorting sub-models based on computation/accuracy in a nested manner. We extend SortedNet to generative NLP tasks, making large language models dynamic without any Pre-Training and by only replacing Standard Fine-Tuning (SFT) with Sorted Fine-Tuning (SoFT). Our approach boosts model efficiency, eliminating the need for multiple models for various scenarios during inference. We show that this approach can unlock the potential of intermediate layers of transformers in generating the target output. Our sub-models remain integral components of the original model, minimizing storage requirements and transition costs between different computational/latency budgets. The efficacy of our proposed method was demonstrated by applying it to tune LLaMA 2 13B on the Stanford Alpaca dataset for instruction following and TriviaQA for closed-book question answering. Our results show the superior performance of sub-models in comparison to Standard Fine-Tuning and SFT+ICT (Early-Exit), all achieved with very efficient tuning and without additional memory usage during inference.

Sorted LLaMA: Unlocking the Potential of Intermediate Layers of Large Language Models for Dynamic Inference

This paper presents a novel \textbf{Cross}-document \textbf{A}bstract \textbf{M}eaning \textbf{R}epresentation (X-AMR) annotation tool designed for annotating key corpus-level event semantics. Leveraging machine assistance through the Prodigy Annotation Tool, we enhance the user experience, ensuring ease and efficiency in the annotation process. Through empirical analyses, we demonstrate the effectiveness of our tool in augmenting an existing event corpus, highlighting its advantages when integrated with GPT-4. Code and annotations

X-AMR Annotation Tool

In this paper we introduce TextBI, a multimodal generic dashboard designed to present multidimensional text annotations on large volumes of multilingual social media data. This tool focuses on four core dimensions: spatial, temporal, thematic, and personal, and also supports additional enrichment data such as sentiment and engagement. Multiple visualization modes are offered, including frequency, movement, and association. This dashboard addresses the challenge of facilitating the interpretation of NLP annotations by visualizing them in a user-friendly, interactive interface catering to two categories of users: (1) domain stakeholders and (2) NLP researchers. We conducted experiments within the domain of tourism leveraging data from X (formerly Twitter) and incorporating requirements from tourism offices. Our approach, TextBI, represents a significant advancement in the field of visualizing NLP annotations by integrating and blending features from a variety of Business Intelligence, Geographical Information Systems and NLP tools. 


Interpreting Predictive Probabilities: Model Confidence or Human Label Variation?

Downloads

Next from EACL 2024

A Comprehensive Evaluation of Inductive Reasoning Capabilities and Problem Solving in Large Language Models

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES