India

The output of large language models (LLMs) is unstable, due both to non-determinism of the decoding process as well as to prompt brittleness. While the intrinsic non-determinism of LLM generation may mimic existing uncertainty in human annotations through distributional shifts in outputs, it is largely assumed, yet unexplored, that the prompt brittleness effect is unique to LLMs. This raises the question: do human annotators show similar sensitivity to prompt changes? If so, should prompt brittleness in LLMs be considered problematic? One may alternatively hypothesize that prompt brittleness correctly reflects human annotation variances. To fill this research gap, we systematically compare the effects of prompt modifications on LLMs and identical instruction modifications for human annotators, focusing on the question of whether humans are similarly sensitive to prompt perturbations. To study this, we prompt both humans and LLMs for a set of text classification tasks conditioned on prompt variations. Our findings indicate that both humans and LLMs exhibit increased brittleness in response to specific types of prompt modifications, particularly those involving the substitution of alternative label sets or label formats. However, the distribution of human judgments is less affected by typographical errors and reversed label order than that of LLMs.

IJCNLP-AACL 2025

Are Humans as Brittle as Large Language Models?

comparison between llm and human

human brittleness

prompt brittleness

### 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!

NAACL 2025

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

technical paper

### 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.

Recent approaches have shown impressive proficiency in extracting and leveraging parametric knowledge from Large-Language Models (LLMs) and Vision-Language Models (VLMs). In this work, we consider how we can improve the retrieval of videos related to complex real-world events by automatically extracting latent parametric knowledge about those events. We present Q2E: a Query-to-Event decomposition method for zero-shot multilingual text-to-video retrieval, adaptable across datasets, domains, LLMs, or VLMs. Our approach demonstrates that we can enhance the understanding of otherwise overly simplified human queries by decomposing the query using the knowledge embedded in LLMs and VLMs. We additionally show how to apply our approach to both visual and speech-based inputs. To combine this varied multimodal knowledge, we adopt entropy-based fusion scoring for zero-shot fusion. Q2E outperforms the previous SOTA on the MultiVENT dataset by 8 NDCG points, while improving on MSR-VTT and MSVD by 4 and 3 points, respectively, outperforming several existing retrieval methods, including many fine-tuned and SOTA zero-shot approaches. We have released both code and data.

Q2E: Query-to-Event Decomposition for Zero-Shot Multilingual Text-to-Video Retrieval

Vision-Language Models (VLMs) often struggle to balance visual and textual information when summarizing complex multimodal inputs, such as entire TV show episodes. In this paper, we propose a zero-shot video-to-text summarization approach that builds its own screenplay-like representation of an episode, effectively integrating key video moments, dialogue, and character information into a unified document. Unlike previous approaches, we simultaneously generate screenplays and name the characters in zero-shot, using only the audio, video, and transcripts as input. Additionally, we highlight that existing summarization metrics can fail to assess the multimodal content in summaries. To address this, we introduce MFactSum, a multimodal metric that evaluates summaries with respect to both vision and text modalities. Using MFactSum, we evaluate our screenplay summaries on the SummScreen3D dataset, demonstrating superiority against state-of-the-art VLMs such as Gemini 1.5 by generating summaries containing 20% more relevant visual information while requiring 75% less of the video as input.

Integrating Video and Text: A Balanced Approach to Multimodal Summary Generation and Evaluation

We investigate communication emerging in noisy environments with the goal of capturing the impact of message disruption on the emerged protocols. We implement two different noise mechanisms, inspired by the erasure and deletion channels studied in information theory, and simulate a referential game in a neural agent-based model with a variable message length channel. We leverage a stochastic evaluation setting to apply noise only after a message is sampled, which adds ecological validity and allows us to estimate information-theoretic measures of the emerged protocol directly from symbol probabilities. Contrary to our expectations, the emerged protocols do not become more redundant with the presence of noise; instead, we observe that certain levels of noise encourage the sender to produce more compositional messages, although the impact varies depending on the type of noise and input representation.

Noise May Drown Out Words but Foster Compositionality: The Advantage of the Erasure and Deletion Noisy Channels on Emergent Communication

The quality of training data is crucial for the performance of supervised machine learning models. In particular, poor annotation quality and spurious correlations between labels and features in text dataset can significantly degrade model generalization. This problem is especially pronounced in harmful language detection, where prior studies have revealed major deficiencies in existing datasets. In this work, we design and test data selection methods based on learnability measures to improve dataset quality. Using a sexism dataset with counterfactuals designed to avoid spurious correlations, we show that pruning with EL2N and PVI scores can lead to significant performance increases and outperforms submodular and random selection. Our analysis reveals that in presence of label imbalance models rely on dataset shortcuts; especially easy-to-classify sexist instances and hard-to-classify non-sexist instances contain shortcuts. Pruning these instances leads to performances increases. Pruning hard-to-classify instances is in general a promising strategy as well when shortcuts are not present.

Enhancing Training Data Quality through Influence Scores for Generalizable Classification: A Case Study on Sexism Detection

Reference-free metrics such as CLIPScore and PAC-S are increasingly used in vision-language tasks due to their scalability and independence from human-written references. However, their reliability under linguistic, visual, and cultural variation remains underexplored. In this work, we present a systematic audit of CLIPScore and PAC-S using an eight-factor diagnostic framework applied to MS-COCO validation images. Our analysis reveals consistent failure modes across dimensions including object size, content category, syntax, named entities, spatial relations and cultural context. Both metrics penalize captions referencing African (−5.5%, −4.8%) and Arabian (−4.9%, −5.3%) cultures, favor large-object and animal-centric scenes (by 20-30%) and show limited sensitivity to spatial negation and word order. CLIPScore correlates more strongly with syntactic complexity, while PAC-S demonstrates greater robustness to verbosity and named–entity variation highlighting complementary strengths rather than superiority. These findings expose cultural and content bias, weak semantic robustness, and limited compositional understanding. We conclude with design recommendations to improve fairness, scale invariance, and semantic grounding in future reference-free evaluation metrics.

A Diagnostic Framework for Auditing Reference-Free Vision-Language Metrics

The rapid proliferation of frontier model agents promises significant societal advances but also raises concerns about systemic risks arising from unsafe interactions.
Collusion to the disadvantage of others has been identified as a central form of undesirable agent cooperation.
The use of information hiding (steganography) in agent communications could render such collusion practically undetectable.
This underscores the need for investigations into the possibility of such behaviours emerging and the robustness corresponding countermeasures.
To investigate this problem we design two approaches -- a gradient-based reinforcement learning (GBRL) method and an in-context reinforcement learning (ICRL) method -- for reliably eliciting sophisticated LLM-generated linguistic text steganography.
We demonstrate, for the first time, that unintended steganographic collusion in LLMs can arise due to mispecified reward incentives during training.
Additionally, we find that standard mitigations --- both passive oversight of model outputs and active mitigation through communication paraphrasing --- are not fully effective at preventing this steganographic communication.
Our findings imply that (i) emergence of steganographic collusion is a plausible concern that should be monitored and researched, and (ii) preventing emergence may require innovation in mitigation techniques.

Hidden in Plain Text: Emergence & Mitigation of Steganographic Collusion in LLMs

There has been a growing research interest in in-image machine translation, which involves translating texts in images from one language to another. Recent studies continue to explore pipeline-based systems due to its straightforward construction and the consistent improvement of its underlined components. However, the existing implementation for such pipeline often lack extensibility, composability, and support for real-time translation. Therefore, this work introduces {\it \imagetra}---an open-source toolkit designed to facilitate the development of the pipeline-based system of in-image machine translation. The toolkit integrates state-of-the-art open-source models and tools, and is designed with a focus on modularity and efficiency, making it particularly well-suited for real-time translation. The toolkit is released at https://github.com/hour/imagetra.

ImageTra: Real-Time Translation for Texts in Image and Video

Live commentary has the potential of making specific broadcasts such as sports or video games more engaging and interesting to watch for spectators. With the recent popularity rise of online live streaming many new categories have entered the space, like art in its many forms or even software development, however, not all live streamers have the capability to be naturally engaging with the audience. We introduce a live commentator assistant system that can discuss what is visible on screen in real time. Our experimental setting is focused on the use-case of a photo editing live stream. We compare several recent vision language models for commentary generation and text to speech models for spoken output, all on relatively modest consumer hardware configurations.

Real-time Commentator Assistant for Photo Editing Live Streaming

Short-form video platforms like TikTok present unique challenges for misinformation detection due to their multimodal, dynamic, and noisy content. We present ShortCheck, a modular, inference-only pipeline with a user-friendly pipeline that automatically identifies checkworthy short-form videos to help human fact-checkers. The system integrates speech transcription, OCR, object and deepfake detection, video-to-text summarization, and claim verification. ShortCheck is validated by evaluating it on two manually annotated datasets with TikTok videos in a multilingual setting. The pipeline achieves promising results with F1-weighted score over 70%. The demo can be accessed live at \url{http://shortcheck.factiverse.ai}.

ShortCheck: Checkworthiness Detection of Multilingual Short‑Form Videos

Accurate assessment of generated chart quality is crucial for automated document creation and editing across diverse applications like finance, medicine, policy making, and education. Current evaluation approaches suffer from significant limitations: human evaluation is costly and difficult to scale, pixel-based metrics ignore data accuracy, while data-centric measures overlook design quality. Recent multimodal LLM evaluators show promise but exhibit concerning inconsistencies due to prompt sensitivity and subjective biases. Existing metrics fail to evaluate chart quality holistically across visual similarity, semantic alignment, and data fidelity, often producing misleading scores that unfairly penalize good charts while rewarding bad ones. We introduce ChartEval, a novel chart evaluation system that compares generated chart images with ground truth by leveraging scene graph parsing to decompose chart images into hierarchical scene graphs of chart objects, attributes, and relations. Subsequently, it applies graph-based similarity measures to compare candidate chart scene graphs against reference scene graphs for measuring chart quality. We demonstrate that our evaluation approach achieves significantly stronger correlation with human judgments compared to existing metrics like GPT-Score, SSIM, and SCRM using a comprehensive benchmark of 4K chart images paired with generation intents and human quality ratings. We demonstrate the utility of the ChartEval system as a reliable automatic chart quality metric on diverse tasks, including language-guided chart editing, chart reconstruction, and text-to-chart synthesis using both open-source and API-based LLMs.

Downloads

Next from IJCNLP-AACL 2025

Q2E: Query-to-Event Decomposition for Zero-Shot Multilingual Text-to-Video Retrieval

Similar lecture

Towards LLMs Robustness to Changes in Prompt Format Styles