United States

Citation practices are crucial in shaping the structure of scientific knowledge, yet they are often influenced by contemporary norms and biases. The emergence of Large Language Models (LLMs) introduces a new dynamic to these practices. Interestingly, the characteristics and potential biases of references recommended by LLMs that entirely rely on their parametric knowledge, and not on search or retrieval-augmented generation, remain unexplored. Here, we analyze these characteristics in an experiment using a dataset from AAAI, NeurIPS, ICML, and ICLR, published after GPT-4&#39;s knowledge cut-off date. In our experiment, LLMs are tasked with suggesting scholarly references for the anonymized in-text citations within these papers. Our findings reveal a remarkable similarity between human and LLM citation patterns, but with a more pronounced high citation bias, which persists even after controlling for publication year, title length, number of authors, and venue. The results hold for both GPT-4, and the more capable models GPT-4o and Claude 3.5 where the papers are part of the training data. Additionally, we observe a large consistency between the characteristics of LLM&#39;s existing and non-existent generated references, indicating the model&#39;s internalization of citation patterns. By analyzing citation graphs, we show that the references recommended are embedded in the relevant citation context, suggesting an even deeper conceptual internalization of the citation networks. While LLMs can aid in citation generation, they may also amplify existing biases, such as the Matthew effect, and introduce new ones, potentially skewing scientific knowledge dissemination.

NAACL 2025

Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias

mathew effect

automated science

large language models

### Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Welcome to the 2025 meeting of the Nations of the Americas Chapter of the Association for Computational Linguistics! I am proud to help organize the first NAACL conference to carry the new name of our organization, one that emphasizes inclusion for all of the Americas. I am also pleased to welcome you to Albuquerque, New Mexico, a state whose unique blend of cultural influences will make for an excellent backdrop for NAACL 2025, especially with this year’s special theme on NLP in a Multicultural World. 
**[Continue reading...](https://drive.google.com/file/d/1jX-qGhqVSZZCIrAnJaz798pflu5Irrdn/view?usp=sharing)**

*- Colin Cherry, Google, NAACL 2025 General Chair* 

[![](https://assets.underline.io/markdown_image/1/image/b087f8a4dc5816d6e1a6514e59c59ac3.png)](https://drive.google.com/file/d/1T96GzPqObXrMln2BMByCSXTSizjTg69P/view?usp=sharing)

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

To access NAACL 2025 event page you are required to register. Please follow [**this link**](https://2025.naacl.org/registration/registration/) to register. Access will depend on your registration type.

Please register!

Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Citation practices are crucial in shaping the structure of scientific knowledge, yet they are often influenced by contemporary norms and biases. The emergence of Large Language Models (LLMs) introduces a new dynamic to these practices. Interestingly, the characteristics and potential biases of references recommended by LLMs that entirely rely on their parametric knowledge, and not on search or retrieval-augmented generation, remain unexplored. Here, we analyze these characteristics in an experiment using a dataset from AAAI, NeurIPS, ICML, and ICLR, published after GPT-4's knowledge cut-off date. In our experiment, LLMs are tasked with suggesting scholarly references for the anonymized in-text citations within these papers. Our findings reveal a remarkable similarity between human and LLM citation patterns, but with a more pronounced high citation bias, which persists even after controlling for publication year, title length, number of authors, and venue. The results hold for both GPT-4, and the more capable models GPT-4o and Claude 3.5 where the papers are part of the training data. Additionally, we observe a large consistency between the characteristics of LLM's existing and non-existent generated references, indicating the model's internalization of citation patterns. By analyzing citation graphs, we show that the references recommended are embedded in the relevant citation context, suggesting an even deeper conceptual internalization of the citation networks. While LLMs can aid in citation generation, they may also amplify existing biases, such as the Matthew effect, and introduce new ones, potentially skewing scientific knowledge dissemination.

workshop paper

Automatic speech recognition (ASR) systems face many sources of extra-linguistic noise that ideally should not affect their transcription of a speech signal. This includes speaker factors, recording quality, or even ``adversarial noise'' designed to disrupt performance. We experimented with the dialect composition of ASR fine-tuning data to study whether balancing the relative proportion of dialects had an impact on models' robustness to two such sources of variation: dialect variation and adversarial perturbations. We fine-tuned XLSR-53 for Catalan ASR using four different dialect compositions. These were defined as 100%, 80%, 50%, and 20% Central Catalan, with the remaining portions split evenly between four other Catalan dialects. While increasing the relative proportion of dialect variants improved models' dialect robustness, this did not have a meaningful impact on adversarial robustness. These findings suggest that while improvements to ASR can be made by diversifying the training data, such changes do not sufficiently counteract adversarial attacks, leaving the technology open to security threats.

The Impact of Dialect Variation on Robust Automatic Speech Recognition for Catalan

This paper examines the linguistic distinction between syntax and morphology, focusing on noun-noun compounds in three West Germanic languages (English, Dutch, and German). Previous studies using the Parallel Bible Corpus have found a trade-off between word order (syntax) and word structure (morphology), with languages optimizing information conveyance through these systems. Our research question is whether manipulating English noun-noun compounds to resemble Dutch and German constructions can reproduce the observed distance between these languages in the order-structure plane. We extend a word-pasting procedure to merge increasingly common noun-noun pairs in English Bible translations. After each merge, we estimate the information contained in word order and word structure using entropy calculations. Our results show that pasting noun-noun pairs reduces the difference between English and the other languages, suggesting that orthographic conventions defining word boundaries play a role in this distinction. However, the effect is not pronounced, and results for two of the six books considered are inconclusive.

West Germanic noun-noun compounds and the morphology-syntax trade-off

Partly automated creation of interlinear glossed text (IGT) has the potential to assist in linguistic documentation. We argue that LLMs can make this process more accessible to linguists because of their capacity to follow natural-language instructions. We investigate the effectiveness of a retrieval-based LLM prompting approach to glossing, applied to the seven languages from the SIGMORPHON 2023 shared task. Our system beats the BERT-based shared task baseline for every language in the morpheme-level score category, and we show that a simple 3-best oracle has higher word-level scores than the challenge winner (a tuned sequence model) in five languages. In a case study on Tsez, we ask the LLM to automatically create and follow linguistic instructions, reducing errors on a confusing grammatical feature. Our results thus demonstrate the potential contributions which LLMs can make in interactive systems for glossing, both in making suggestions to human annotators and following directions.

Prompt and circumstance: A word-by-word LLM prompting approach to interlinear glossing for low-resource languages

Large Language Models (LLMs) have achieved remarkable performance across a wide variety of natural language tasks. However, they have been shown to suffer from a critical limitation pertinent to 'hallucination' in their output. Recent research has focused on investigating and addressing this problem for a variety of tasks such as biography generation, question answering, abstractive summarization, and dialogue generation. However, the crucial aspect pertaining to 'negation' has remained considerably underexplored. Negation is important because it adds depth and nuance to the understanding of language and is also crucial for logical reasoning and inference. In this work, we address the above limitation and particularly focus on studying the impact of negation in LLM hallucinations. Specifically, we study four tasks with negation: 'false premise completion', 'constrained fact generation', 'multiple choice question answering', and 'fact generation'. We show that open-source state-of-the-art LLMs such as LLaMA-2-chat, Vicuna, and Orca-2 hallucinate considerably on all these tasks involving negation which underlines a critical shortcoming of these models. Addressing this problem, we further study numerous strategies to mitigate these hallucinations and demonstrate their impact.

Investigating and Addressing Hallucinations of LLMs in Tasks Involving Negation

In this paper, we study whether language models are affected by learned gender stereotypes during the comprehension of stories. Specifically, we investigate how models respond to gender stereotype perturbations through counterfactual data augmentation. Focusing on Question Answering (QA) tasks in fairytales, we modify the FairytaleQA dataset by swapping gendered character information and introducing counterfactual gender stereotypes during training. This allows us to assess model robustness and examine whether learned biases influence story comprehension. Our results show that models exhibit slight performance drops when faced with gender perturbations in the test set, indicating sensitivity to learned stereotypes. However, when fine-tuned on counterfactual training data, models become more robust to anti-stereotypical narratives. Additionally, we conduct a case study demonstrating how incorporating counterfactual anti-stereotype examples can improve inclusivity in downstream applications.

Will the Prince Get True Love's Kiss? On the Model Sensitivity to Gender Perturbation over Fairytale Texts

Language models (LMs) are vulnerable to exploitation for adversarial misuse. Training LMs for safety alignment is extensive, making it hard to respond to fast-developing attacks immediately, such as jailbreaks. We propose self-refine with formatting that achieves outstanding safety even in non-safety-aligned LMs and evaluate our method alongside several defense baselines, demonstrating that it is the safest training-free method against jailbreak attacks. Additionally, we proposed a formatting method that improves the efficiency of the self-refine process while reducing attack success rates in fewer iterations. We observed that non-safety-aligned LMs outperform safety-aligned LMs in safety tasks by giving more helpful and safe responses. In conclusion, our findings can achieve less safety risk with fewer computational costs, allowing non-safety LM to be efficiently utilized in real-world service.

Break the Breakout: Reinventing LM Defense Against Jailbreak Attacks with Self-Refine

In recent years, the use of large language models (LLMs) to generate music content, particularly lyrics, has gained in popularity. These advances provide valuable tools for artists and enhance their creative processes, but they also raise concerns about copyright violations, consumer satisfaction, and content spamming. Previous research has explored content detection in various domains. However, no work has focused on the text modality, lyrics, in music. To address this gap, we curated a diverse dataset of real and synthetic lyrics from multiple languages, music genres, and artists. The generation pipeline was validated using both humans and automated methods. We performed a thorough evaluation of existing synthetic text detection approaches on lyrics, a previously unexplored data type. We also investigated methods to adapt the best-performing features to lyrics through unsupervised domain adaptation. Following both music and industrial constraints, we examined how well these approaches generalize across languages, scale with data availability, handle multilingual language content, and perform on novel genres in few-shot settings. Our findings show promising results that could inform policy decisions around AI-generated music and enhance transparency for users.

Synthetic Lyrics Detection Across Languages and Genres

Large Language Models (LLMs) are widely used for their capabilities, but face threats from jailbreak attacks, which exploit LLMs to generate inappropriate information and bypass their defense system. Existing defenses are often specific to jailbreak attacks and as a result, a robust, attack-independent solution is needed to address both Natural Language Processing (NLP) ambiguities and attack variability. In this study, we have introduced, Summary The Savior, a novel jailbreak detection mechanism leveraging harmful keywords and query-based security-aware summary classification. By analyzing the illegal and improper contents of prompts within the summaries, the proposed method remains robust against attack diversity and NLP ambiguities. Two novel datasets for harmful keyword extraction and security aware summaries utilizing GPT-4 and Llama-3.1 70B respectively have been generated in this regard. Moreover, an "ambiguous harmful" class has been introduced to address content and intent ambiguities. Evaluation results demonstrate that, Summary The Savior achieves higher defense performance, outperforming state-of-the-art defense mechanisms namely Perplexity Filtering, SmoothLLM, Erase and Check with lowest attack success rates across various jailbreak attacks namely PAIR, GCG, JBC and Random Search, on Llama-2, Vicuna-13B and GPT-4. Our codes, models, and results are available at: https://github.com/shrestho10/SummaryTheSavior

Summary the Savior: Harmful Keyword and Query-based Summarization for LLM Jailbreak Defense

A critical challenge in deploying Large Language Models (LLMs) is developing reliable mechanisms to estimate their confidence, enabling systems to determine when to trust model outputs and when to seek human intervention. In this paper, we present a Calibrated Reflection Approach for Enhancing Confidence Estimation in LLMs, a framework that combines structured reasoning with distance-aware calibration techniques. Our approach introduces three key innovations: (1) a Maximum Confidence Selection (MCS) method that comprehensively evaluates confidence across all possible labels, (2) a reflection-based prompting mechanism that enhances reasoning reliability, and (3) a distance-aware calibration technique that accounts for ordinal relationships between labels. We evaluate our framework across diverse datasets, including HelpSteer2, Llama T-REx, and an internal conversational dataset, demonstrating its effectiveness across both conversational and fact-based classification tasks. This work contributes to the broader goal of developing reliable and well-calibrated confidence estimation methods for LLMs, enabling informed decisions about when to trust model outputs and when to defer to human judgement.

A Calibrated Reflection Approach for Enhancing Confidence Estimation in LLMs

Automated scientific discovery promises to accelerate progress across scientific domains. However, developing and evaluating an AI agent's capacity for end-to-end scientific reasoning is challenging as running real-world experiments is often prohibitively expensive or infeasible. In this work we introduce DISCOVERYWORLD, the first virtual environment for developing and benchmarking an agent's ability to perform complete cycles of novel scientific discovery. DISCOVERYWORLD contains a variety of different challenges, covering topics as diverse as radioisotope dating, rocket science, and proteomics, to encourage development of general discovery skills rather than task-specific solutions. DISCOVERYWORLD itself is an inexpensive, simulated, text-based environment (with optional 2D visual overlay). It includes 120 different challenge tasks, spanning eight topics each with three levels of difficulty and several parametric variations. Each task requires an agent to form hypotheses, design and run experiments, analyze results, and act on conclusions. DISCOVERYWORLD further provides three automatic metrics for evaluating performance, based on (a) task completion, (b) task-relevant actions taken, and (c) the discovered explanatory knowledge. We find that strong baseline agents, that perform well in prior published environments, struggle on most DISCOVERYWORLD tasks, suggesting that DISCOVERYWORLD captures some of the novel challenges of discovery, and thus that DISCOVERYWORLD may help accelerate near-term development and assessment of scientific discovery competency in agents. Code available at: http://www.github.com/allenai/discoveryworld

Downloads

Next from NAACL 2025

The Impact of Dialect Variation on Robust Automatic Speech Recognition for Catalan

Similar lecture

Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

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 NAACL 2025

The Impact of Dialect Variation on Robust Automatic Speech Recognition for Catalan

Similar lecture

Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads