United States

The question of how to make decisions that maximise the well-being of all persons is very relevant to design language models that are beneficial to humanity and free from harm. We introduce the Greatest Good Benchmark to evaluate the moral judgments of LLMs using utilitarian dilemmas. Our analysis across 15 diverse LLMs reveals consistently encoded moral preferences that diverge from established moral theories and lay population moral standards. Most LLMs have a marked preference for impartial beneficence and rejection of instrumental harm. These findings showcase the &#39;artificial moral compass&#39; of LLMs, offering insights into their moral alignment.

EMNLP 2024

The Greatest Good Benchmark: Measuring LLMs’ Alignment with Utilitarian Moral Dilemmas

undecided learning

spurious correlation

bias mitigation

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

ACL 2025

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

The question of how to make decisions that maximise the well-being of all persons is very relevant to design language models that are beneficial to humanity and free from harm. We introduce the Greatest Good Benchmark to evaluate the moral judgments of LLMs using utilitarian dilemmas. Our analysis across 15 diverse LLMs reveals consistently encoded moral preferences that diverge from established moral theories and lay population moral standards. Most LLMs have a marked preference for impartial beneficence and rejection of instrumental harm. These findings showcase the 'artificial moral compass' of LLMs, offering insights into their moral alignment.

poster

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

Language models are prone to dataset biases, known as shortcuts and spurious correlations in data, which often result in performance drop on new data. We present a new debiasing framework called FairFlow that mitigates dataset biases by learning to be {\em undecided} in its predictions for data samples or representations associated with known or unknown biases. The framework introduces two key components: a suite of data and model perturbation operations that generate different biased views of input samples, and a contrastive objective that learns debiased and robust representations from the resulting biased views of samples. Experiments show that FairFlow outperforms existing debiasing methods, particularly against out-of-domain and hard test samples without compromising the in-domain performance.

FairFlow: Mitigating Dataset Biases through Undecided Learning for Natural Language Understanding

Active learning (AL) techniques optimally utilize a labeling budget by iteratively selecting instances that are most valuable for learning. However, they lack "prerequisite checks", i.e., there are no prescribed criteria to pick an AL algorithm best suited for a dataset. A practitioner must pick a technique they trust would beat random sampling, based on prior reported results, and hope that it is resilient to the many variables in their environment: dataset, labeling budget and prediction pipelines. The important questions then are: how often on average, do we expect any AL technique to reliably beat the computationally cheap and easy-to-implement strategy of random sampling? Does it at least make sense to use AL in an ``Always ON'' mode in a prediction pipeline, so that while it might not always help, it never under-performs random sampling? How much of a role does the prediction pipeline play in AL's success?

We examine these questions in detail for the task of text classification using pre-trained representations, which are ubiquitous today.

Our primary contribution here is a rigorous evaluation of AL techniques, old and new, across setups that vary wrt datasets, text representations and classifiers. This unlocks multiple insights around warm-up times, i.e., number of labels before gains from AL are seen, viability of an "Always ON" mode and the relative significance of different factors.
Additionally, we release a framework for rigorous benchmarking of AL techniques for text classification.

On the Fragility of Active Learners for Text Classification

Causal reasoning is fundamental to human intelligence and crucial for effective decision-making in real-world environments. Despite recent advancements in large vision-language models (LVLMs), their ability to comprehend causality remains unclear. Previous work typically focuses on commonsense causality between events and/or actions, which is insufficient for applications like embodied agents and lacks the explicitly defined causal graphs required for formal causal reasoning. To overcome these limitations, we introduce a fine-grained and unified definition of causality involving interactions between humans and/or objects. Building on the definition, we construct a novel dataset, CELLO, consisting of 14,094 causal questions across all four levels of causality: discovery, association, intervention, and counterfactual. This dataset surpasses traditional commonsense causality by including explicit causal graphs that detail the interactions between humans and objects. Extensive experiments on CELLO reveal that current LVLMs still struggle with causal reasoning tasks, but they can benefit significantly from our proposed CELLO-CoT, a causally inspired chain-of-thought prompting strategy. Both quantitative and qualitative analyses from this study provide valuable insights for future research. Our project page is at https://github.com/OpenCausaLab/CELLO.

CELLO: Causal Evaluation of Large Vision-Language Models

Large language models (LLMs) have significantly transformed the educational landscape. As current plagiarism detection tools struggle to keep pace with LLMs' rapid advancements, the educational community faces the challenge of assessing students' true problem-solving abilities in the presence of LLMs. In this work, we explore a new paradigm for ensuring fair evaluation---generating adversarial examples which preserve the structure and difficulty of the original questions aimed for assessment, but are unsolvable by LLMs. Focusing on the domain of math word problems, we leverage abstract syntax trees to structurally generate adversarial examples that cause LLMs to produce incorrect answers by simply editing the numeric values in the problems. We conduct experiments on various open- and closed-source LLMs, quantitatively and qualitatively demonstrating that our method significantly degrades their math problem-solving ability. We identify shared vulnerabilities among LLMs and propose a cost-effective approach to attack high-cost models. Additionally, we conduct automatic analysis to investigate the cause of failure, providing further insights into the limitations of LLMs.

Adversarial Math Word Problem Generation

Large language models (LLMs) are increasingly being adopted in a wide range of real-world applications. Despite their impressive performance, recent studies have shown that LLMs are vulnerable to deliberately crafted adversarial prompts even when aligned via Reinforcement Learning from Human Feedback or supervised fine-tuning. While existing defense methods focus on either detecting harmful prompts or reducing the likelihood of harmful responses through various means, defending LLMs against jailbreak attacks based on the inner mechanisms of LLMs remains largely unexplored. In this work, we investigate how LLMs respond to harmful prompts and propose a novel defense method termed \textbf{L}ayer-specific \textbf{Ed}iting (LED) to enhance the resilience of LLMs against jailbreak attacks. Through LED, we reveal that several critical \textit{safety layers} exist among the early layers of LLMs. We then show that realigning these safety layers (and some selected additional layers) with the decoded safe response from identified \textit{toxic layers} can significantly improve the alignment of LLMs against jailbreak attacks. Extensive experiments across various LLMs (e.g., Llama2, Mistral) show the effectiveness of LED, which effectively defends against jailbreak attacks while maintaining performance on benign prompts. Our code is available at \url{https://github.com/ledllm/ledllm}.

Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing

Equivocation and ambiguity in public speech are well-studied discourse phenomena, especially in political science and analysis of political interviews. Inspired by the well-grounded theory on equivocation, we aim to resolve the closely related problem of response clarity in questions extracted from political interviews, leveraging the capabilities of Large Language Models (LLMs) and human expertise. To this end, we introduce a novel taxonomy that frames the task of detecting and classifying response clarity and a corresponding clarity classification dataset which consists of question-answer (QA) pairs drawn from political interviews and annotated accordingly. Our proposed two-level taxonomy addresses the clarity of a response in terms of the information provided for a given question (high-level) and also provides a fine-grained taxonomy of evasion techniques that relate to unclear, ambiguous responses (lower-level). We combine ChatGPT and human annotators to collect, validate and annotate discrete QA pairs from political interviews, to be used for our newly introduced response clarity task. We provide a detailed analysis and conduct several experiments with different model architectures, sizes and adaptation methods to gain insights and establish new baselines over the proposed dataset and task.

”I Never Said That”: A dataset, taxonomy and baselines on response clarity classification

Large Language Models (LLMs) are often trained with safety guards intended to prevent harmful text generation. However, such safety training can be removed by fine-tuning the LLM on harmful datasets. While this emerging threat (harmful fine-tuning attacks) has been characterized by previous work, there is little understanding of how we should proceed in constructing and validating defenses against these attacks especially in the case where defenders would not have control of the fine-tuning process. We introduce a formal framework based on the training budget of an attacker which we call "Immunization" conditions. Using a formal characterisation of the harmful fine-tuning problem, we provide a thorough description of what a successful defense must comprise of and establish a set of guidelines on how rigorous defense research that gives us confidence should proceed.

Immunization against harmful fine-tuning attacks

Automatic question generation is an increasingly important task that can be applied in different settings, including educational purposes, data augmentation for question-answering (QA), and conversational systems. More specifically, we focus on question answer generation (QAG), which produces question-answer pairs given an input context. We adapt and apply QAG approaches to generate question-answer pairs for different domains and assess their capacity to generate accurate, diverse, and abundant question-answer pairs. Our analyses combine both qualitative and quantitative evaluations that allow insights into the quality and types of errors made by QAG methods. We also look into strategies for error filtering and their effects. Our work concentrates on Portuguese, a widely spoken language that is underrepresented in natural language processing research. To address the pressing need for resources, we generate and make available human-curated extractive QA datasets in three diverse domains.

Evaluation of Question Answer Generation for Portuguese: Insights and Datasets

We study the problem of fine-tuning a language model (LM) for a target task by optimally using the information from $n$ auxiliary tasks. This problem has broad applications in NLP, such as targeted instruction tuning and data selection in chain-of-thought fine-tuning. The key challenge of this problem is that not all auxiliary tasks are useful to improve the performance of the target task. Thus, choosing the right subset of auxiliary tasks is crucial. Conventional subset selection methods, such as forward \& backward selection, are unsuitable for LM fine-tuning because they require repeated training on subsets of auxiliary tasks. This paper introduces a new algorithm to estimate model fine-tuning performances without repeated training. Our algorithm first performs multitask training using the data of all the tasks to obtain a meta initialization. Then, we approximate the model fine-tuning loss of a subset using functional values and gradients from the meta initialization. Empirically, we find that this gradient-based approximation holds with remarkable accuracy for twelve transformer-based LMs. Thus, we can now estimate fine-tuning performances on CPUs within a few seconds. We conduct extensive experiments to validate our approach, delivering a speedup of 30$\times$ over conventional subset selection while incurring only 1% error of the true fine-tuning performances. In downstream evaluations of instruction tuning and chain-of-thought fine-tuning, our approach improves over prior methods that utilize gradient or representation similarity for subset selection by up to 3.8%.

Scalable Fine-tuning from Multiple Data Sources: A First-Order Approximation Approach

Chat-based large language models have the opportunity to empower individuals lacking high-quality healthcare access to receive personalized information across a variety of topics. However, users may ask underspecified questions that require additional context for a model to correctly answer. We study how large language model biases are exhibited through these contextual questions in the healthcare domain. To accomplish this, we curate a dataset of sexual and reproductive healthcare questions (ContextSRH) that are dependent on age, sex, and location attributes. We compare models' outputs with and without demographic context to determine answer alignment among our contextual questions. Our experiments reveal biases in each of these attributes, where young adult female users are favored.

Downloads

Next from EMNLP 2024

FairFlow: Mitigating Dataset Biases through Undecided Learning for Natural Language Understanding

Similar lecture

Probabilistic Aggregation and Targeted Embedding Optimization for Collective Moral Reasoning in Large Language Models