India

Large language models have the potential to generate explanations for their own predictions in a variety of styles based on user instructions. Recent research has examined whether these self-explanations faithfully reflect the models&#39; actual behavior and has found that they often lack faithfulness. However, the question of how to improve faithfulness remains underexplored. Moreover, because different explanation styles have superficially distinct characteristics, it is unclear whether improvements observed in one style also arise when using other styles. This study analyzes the effects of training for faithful self-explanations and the extent to which these effects generalize, using three classification tasks and three explanation styles. We construct one-word constrained explanations that are likely to be faithful using a feature attribution method, and use these pseudo-faithful self-explanations for continual learning on instruction-tuned models. Our experiments demonstrate that training can improve self-explanation faithfulness across all classification tasks and explanation styles, and that these improvements also show signs of generalization to the multi-word settings and to unseen tasks. Furthermore, we find consistent cross-style generalization among three styles, suggesting that training may contribute to a broader improvement in faithful self-explanation ability.

IJCNLP-AACL 2025

Investigating Training and Generalization in Faithful Self-Explanations of Large Language Models

explanation faithfulness

free-text/natural language explanations

continual learning

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

Large language models have the potential to generate explanations for their own predictions in a variety of styles based on user instructions. Recent research has examined whether these self-explanations faithfully reflect the models' actual behavior and has found that they often lack faithfulness. However, the question of how to improve faithfulness remains underexplored. Moreover, because different explanation styles have superficially distinct characteristics, it is unclear whether improvements observed in one style also arise when using other styles. This study analyzes the effects of training for faithful self-explanations and the extent to which these effects generalize, using three classification tasks and three explanation styles. We construct one-word constrained explanations that are likely to be faithful using a feature attribution method, and use these pseudo-faithful self-explanations for continual learning on instruction-tuned models. Our experiments demonstrate that training can improve self-explanation faithfulness across all classification tasks and explanation styles, and that these improvements also show signs of generalization to the multi-word settings and to unseen tasks. Furthermore, we find consistent cross-style generalization among three styles, suggesting that training may contribute to a broader improvement in faithful self-explanation ability.

poster

### Welcome to IJCNLP-AACL 2025! 
 It is a great honor to host this joint conference in Mumbai, India, from December 20 to 24, 2025. The joint conferences of IJCNLP and AACL are organized with alternating leadership in the Asia-Pacific region. The event is run by the Asian Federation of Natural Language Processing (AFNLP) in odd years, and by AACL in even years, while it is organized solely by ACL when the annual ACL meeting is held in the region. This year, the conference is primarily organized by AFNLP. 
*Kentaro Inui
MBZUAI, UAE
General Chair, IJCNLP-AACL 2025* 
Read full message and download the Conference Handbook [**here**](https://drive.google.com/file/d/1UTwxkAqSqI-GAoJC3wE1zZt5VP1Y8GX0/view?usp=sharing).

The 14th IJCNLP & 4th AACL will be held in Mumbai, India from December 20th to December 24th, 2025.

Clinical interventions often hinge on age: medications and procedures safe for adults may be harmful to children or ineffective for older adults. However, as language models are increasingly integrated into biomedical evidence synthesis workflows, it remains uncertain whether these systems preserve such crucial demographic distinctions. To address this gap, we evaluate how well state-of-the-art language models retain age-related information when generating abstractive summaries of biomedical studies.

We construct DemogSummary, a novel age-stratified dataset of systematic review primary studies, covering child, adult, and older adult populations. We evaluate three prominent summarisation-capable LLMs, Qwen (open-source), Longformer (open-source) and GPT-4.1 Nano (proprietary), using both standard metrics and a newly proposed Demographic Salience Score (DSS), which quantifies age-related entity retention and hallucination.

Our results reveal systematic disparities across models and age groups: demographic fidelity is lowest for adult-focused summaries, and underrepresented populations are more prone to hallucinations. These findings highlight the limitations of current LLMs in faithful and bias-free summarisation and point to the need for fairness-aware evaluation frameworks and summarisation pipelines in biomedical NLP.

LLMs Do Not See Age: Assessing Demographic Bias in Automated Systematic Review Synthesis

While Large Language Models (LLMs) have shown remarkable performance in various Natural Language Processing (NLP) tasks, their effectiveness seem to be heavily biased toward high-resource languages. This proposal aims to address this gap by developing efficient training strategies for low-resource languages. We propose various techniques for efficient learning in simluated low-resource settings for English. We then plan to adapt these methods for low-resource languages. We plan to experiment with both natural language generation and understanding models. We evaluate the models on similar benchmarks as the BabyLM challenge for English. For other languages, we plan to use treebanks and translation techniques to create our own silver test set to evaluate the low-resource LMs.

Thesis Proposal: Efficient Methods for Natural Language Generation/Understanding Systems

Most models struggle with deception detection in diplomatic conversations due to extreme class imbalance and a lack of "social awareness" in how speakers and listeners are represented. To our end, we present CORAL (COntextual Representation and Asymmetric Learning), a model that learns socially-aware representations—capturing both speaker behavioral traits and their power differentials (social hierarchy)—and couples them with a principled learning framework designed to handle deception’s rarity. CORAL constructs embeddings that capture two key social signals: stable speaker traits through learned signatures and dynamic interaction contexts through power differentials. These embeddings form the input to a classifier trained with a Positive-Unlabeled (PU) learning objective, designed for the asymmetric nature of deception (rare lies versus abundant, unlabeled truths). By synergizing representation and learning, CORAL achieves a state-of-the-art Macro F1-score of 0.65 on the highly imbalanced Peskov Diplomacy dataset, surpassing models that tackle these challenges in isolation. Our findings highlight that advancing socially complex NLP tasks requires aligning richer representations with objectives tailored to data asymmetry.

CORAL: A Synergistic Approach to Deception Detection with Socially-Aware Representations and Positive-Unlabeled Learning

The relationship between tokenizer algorithm (e.g., Byte-Pair Encoding (BPE), Unigram), morphological alignment, tokenization quality (e.g., compression efficiency), and downstream performance remains largely unclear, particularly for languages with complex morphology. In this paper, we conduct a comprehensive evaluation of tokenizers using small-sized BERT models---from pre-training through fine-tuning---for Telugu (agglutinative), along with preliminary evaluation in Hindi (primarily fusional with some agglutination) and English (fusional). To evaluate morphological alignment of tokenizers in Telugu, we create a dataset containing gold morpheme segmentations of 600 derivational and 7000 inflectional word forms.

Our experiments reveal two key findings for Telugu. First, the choice of tokenizer algorithm is the most significant factor influencing performance, with Unigram-based tokenizers consistently outperforming BPE across most settings. Second, while better morphological alignment shows a moderate, positive correlation with performance on text classification and structure prediction tasks, its impact is secondary to the tokenizer algorithm. Notably, hybrid approaches that use morphological information for pre-segmentation significantly boost the performance of BPE, though not Unigram. Our results further showcase the need for comprehensive intrinsic evaluation metrics for tokenizers that could explain downstream performance trends consistently.

Rethinking Tokenization for Rich Morphology: The Dominance of Unigram over BPE and Morphological Alignment

Semantic role labeling (SRL) is a fundamental task in natural language processing that is crucial for achieving deep semantic understanding. Despite the success of large language models (LLMs) in several downstream NLP tasks, key tasks such as SRL remain a challenge for LLMs. Hence, in this study, we attempt to instantiate the efficacy of LLMs for the task of SRL via Question answering. Toward that goal, we investigate the effectiveness of five different LLMs (Llama, Mistral, Qwen, OpenChat, Gemini) using zero-shot and few-shot prompting. Our findings indicate that few-shot prompting enhances the performance of all models. Although Gemini outperformed others by a margin of 11%, Qwen and Llama are not too far behind. Additionally, we conduct a comprehensive error analysis to shed light on the cases where LLMs fail. This study offers valuable insights into the performance of LLMs for structured prediction and the effectiveness of simple prompting techniques in the Question-Answering framework for SRL.

Are LLMS Good for Semantic Role Labeling via Question Answering?: A Preliminary Analysis

Sarcasm is a specific form of ironic speech which can often be hard to understand for language models due to its nuanced nature. Recent improvements in the ability of such models to detect and generate sarcasm motivate us to try a new approach to help language models perceive sarcasm as a speech style, through a human cognitive perspective. In this work, we propose a multi-hop Chain of Thought (CoT) methodology to understand the context of an utterance that follows a dialogue and to perform bidirectional style transfer on that utterance, leveraging the Theory of Mind. We use small language models (SLMs) due to their cost-efficiency and fast response-time. The generated utterances are evaluated using both LLM-as-a-judge and human evaluation, suitable to the open-ended and stylistic nature of the generations. Along with these, we also evaluate scores of automated metrics such as DialogRPT, BLEU and SBERT; drawing valuable insights from them that support our evidence. Based on this, we find that our cognitive approach to sarcasm is an effective way for language models to stylistically understand and generate sarcasm with better authenticity.

Could you BE more sarcastic? A Cognitive Approach to Bidirectional Sarcasm Understanding in Language Models

Large Language Models (LLMs) often hallucinate, generating nonsensical or false information that can be especially harmful in sensitive fields such as medicine or law. To study this phenomenon systematically, we introduce FalseCite, a curated dataset designed to capture and benchmark hallucinated responses induced by misleading or fabricated citations. Running GPT-4o-mini, Falcon-7B, and Mistral 7-B through FalseCite, we observed a noticeable increase in hallucination activity for false claims with deceptive citations, especially in GPT-4o-mini. Using the responses from FalseCite, we can also analyze the internal states of hallucinating models, visualizing and clustering the hidden state vectors. From this analysis, we noticed that the hidden state vectors, regardless of hallucination or non-hallucination, tend to trace out a distinct horn-like shape. Our work underscores FalseCite’s potential as a foundation for evaluating and mitigating hallucinations in future LLM research.

Visualizing and Benchmarking LLM Factual Hallucination Tendencies via Internal State Analysis and Clustering

Emotional prompting—the use of specific emotional diction in prompt engineering—has shown increasing promise in improving large language model (LLM) performance, truthfulness, and responsibility, however these studies have been limited to single types of positive emotional stimuli and have not considered varying degrees of emotion intensity in their analyses. In this paper, we explore the effects of four distinct emotions—joy, encouragement, anger, and insecurity—in emotional prompting and evaluate them on accuracy, sycophancy, and toxicity. We develop a prompt-generation pipeline with GPT-4o mini to create a suite of LLM and human-generated prompts with varying intensities across the four emotions. Then, we compile a "Gold Dataset" of prompts where human and model labels align. Our empirical evaluation on LLM behavior suggests that positive emotional stimuli lead to more accurate and less toxic results, but also increase sycophantic behavior.

The Role of Emotional Stimuli and Intensity in Shaping Large Language Model Behavior

As neural networks are increasingly deployed in dynamic environments, they face the challenge of catastrophic forgetting, the tendency to overwrite previously learned knowledge when adapting to new tasks, resulting in severe performance degradation on earlier tasks. We propose Selective Forgetting-Aware Optimization (SFAO), a dynamic method that regulates gradient directions via cosine similarity and per-layer gating, enabling controlled forgetting while balancing plasticity and stability. SFAO selectively projects, accepts, or discards updates using a tunable mechanism with efficient Monte Carlo approximation. Experiments on standard continual learning benchmarks show that SFAO achieves competitive accuracy with markedly lower memory cost, a 90% reduction, and improved forgetting on MNIST datasets, making it suitable for resource-constrained scenarios.

Mitigating Forgetting in Continual Learning with Selective Gradient Projection

We demonstrate that self-generated examples enhance LLM reasoning mainly through the creation process, rather than their content.

Downloads

Next from IJCNLP-AACL 2025

LLMs Do Not See Age: Assessing Demographic Bias in Automated Systematic Review Synthesis

Similar lecture

Are self-explanations from Large Language Models faithful?

Stay up to date with the latest Underline news!

PRODUCT

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from IJCNLP-AACL 2025

LLMs Do Not See Age: Assessing Demographic Bias in Automated Systematic Review Synthesis

Similar lecture

Are self-explanations from Large Language Models faithful?

Stay up to date with the latest Underline news!

PRODUCT

COMPANY

RESOURCES

Downloads