HapticVLM: VLM-Driven Texture Recognition Aimed at Intelligent Haptic Interaction

This paper introduces HapticVLM, a novel multimodal system that integrates vision-language reasoning with deep convolutional networks to enable real-time haptic feedback. HapticVLM leverages a ConvNeXt-based material recognition module to generate robust visual embeddings for accurate identification of object materials, while a state-of-the-art Vision-Language Model (Qwen2-VL-2B-Instruct) infers ambient temperature from environmental cues. The system synthesizes tactile sensations by delivering vibrotactile feedback through speakers and thermal cues via a Peltier module, thereby bridging the gap between visual perception and tactile experience. Experimental evaluations demonstrate an average recognition accuracy of 84.67% across five distinct auditory-tactile patterns and a temperature estimation accuracy of 86.7% based on a tolerance-based evaluation method with an 8{\deg}C margin of error across 15 scenarios. Although promising, the current study is limited by the use of a small set of prominent patterns and a modest participant pool. Future work will focus on expanding the range of tactile patterns and increasing user studies to further refine and validate the system's performance. Overall, HapticVLM presents a significant step toward context-aware, multimodal haptic interaction with potential applications in virtual reality, and assistive technologies.

Improving Informally Romanized Language Identification

The Latin script is often used to informally write languages with non-Latin native scripts. In many cases (e.g., most languages in India), there is no conventional spelling of words in the Latin script, hence there will be high spelling variability in written text. Such romanization renders languages that are normally easily distinguished based on script highly confusable, such as Hindi and Urdu. In this work, we increase language identification (LID) accuracy for romanized text by improving the methods used to synthesize training sets. We find that training on synthetic samples which incorporate natural spelling variation yields higher LID system accuracy than including available naturally occurring examples in the training set, or even training higher capacity models. We demonstrate new state-of-the-art LID performance on romanized text from 20 Indic languages in the Bhasha-Abhijnaanam evaluation set (Madhani et al., 2023a), improving test F1 from the reported 74.7% (using a pretrained neural model) to 85.4% using a linear classifier trained solely on synthetic data and 88.2% when also training on available harvested text.

VTS-LLM: Domain-Adaptive LLM Agent for Enhancing Awareness in Vessel Traffic Services through Natural Language

Vessel Traffic Services (VTS) are essential for maritime safety and regulatory compliance through real-time traffic management. However, with increasing traffic complexity and the prevalence of heterogeneous, multimodal data, existing VTS systems face limitations in spatiotemporal reasoning and intuitive human interaction. In this work, we propose VTS-LLM Agent, the first domain-adaptive large LLM agent tailored for interactive decision support in VTS operations. We formalize risk-prone vessel identification as a knowledge-augmented Text-to-SQL task, combining structured vessel databases with external maritime knowledge. To support this, we construct a curated benchmark dataset consisting of a custom schema, domain-specific corpus, and a query-SQL test set in multiple linguistic styles. Our framework incorporates NER-based relational reasoning, agent-based domain knowledge injection, semantic algebra intermediate representation, and query rethink mechanisms to enhance domain grounding and context-aware understanding. Experimental results show that VTS-LLM outperforms both general-purpose and SQL-focused baselines under command-style, operational-style, and formal natural language queries, respectively. Moreover, our analysis provides the first empirical evidence that linguistic style variation introduces systematic performance challenges in Text-to-SQL modeling. This work lays the foundation for natural language interfaces in vessel traffic services and opens new opportunities for proactive, LLM-driven maritime real-time traffic management.

Red Teaming Large Language Models for Healthcare

We present the design process and findings of the pre-conference workshop at the Machine Learning for Healthcare Conference (2024) entitled Red Teaming Large Language Models for Healthcare, which took place on August 15, 2024. Conference participants, comprising a mix of computational and clinical expertise, attempted to discover vulnerabilities -- realistic clinical prompts for which a large language model (LLM) outputs a response that could cause clinical harm. Red-teaming with clinicians enables the identification of LLM vulnerabilities that may not be recognised by LLM developers lacking clinical expertise. We report the vulnerabilities found, categorise them, and present the results of a replication study assessing the vulnerabilities across all LLMs provided.

Retrieval-Enhanced Few-Shot Prompting for Speech Event Extraction

Speech Event Extraction (SpeechEE) is a challenging task that lies at the intersection of Automatic Speech Recognition (ASR) and Natural Language Processing (NLP), requiring the identification of structured event information from spoken language. In this work, we present a modular, pipeline-based SpeechEE framework that integrates high-performance ASR with semantic search-enhanced prompting of Large Language Models (LLMs). Our system first classifies speech segments likely to contain events using a hybrid filtering mechanism including rule-based, BERT-based, and LLM-based models. It then employs few-shot LLM prompting, dynamically enriched via semantic similarity retrieval, to identify event triggers and extract corresponding arguments. We evaluate the pipeline using multiple LLMs (Llama3-8B, GPT-4o-mini, and o1-mini) highlighting significant performance gains with o1-mini, which achieves 63.3% F1 on trigger classification and 27.8% F1 on argument classification, outperforming prior benchmarks. Our results demonstrate that pipeline approaches, when empowered by retrieval-augmented LLMs, can rival or exceed end-to-end systems while maintaining interpretability and modularity. This work provides practical insights into LLM-driven event extraction and opens pathways for future hybrid models combining textual and acoustic features.

Simple Visual Artifact Detection in Sora-Generated Videos

The December 2024 release of OpenAI's Sora, a powerful video generation model driven by natural language prompts, highlights a growing convergence between large language models (LLMs) and video synthesis. As these multimodal systems evolve into video-enabled LLMs (VidLLMs), capable of interpreting, generating, and interacting with visual content, understanding their limitations and ensuring their safe deployment becomes essential. This study investigates visual artifacts frequently found and reported in Sora-generated videos, which can compromise quality, mislead viewers, or propagate disinformation. We propose a multi-label classification framework targeting four common artifact label types: label 1: boundary / edge defects, label 2: texture / noise issues, label 3: movement / joint anomalies, and label 4: object mismatches / disappearances. Using a dataset of 300 manually annotated frames extracted from 15 Sora-generated videos, we trained multiple 2D CNN architectures (ResNet-50, EfficientNet-B3 / B4, ViT-Base). The best-performing model trained by ResNet-50 achieved an average multi-label classification accuracy of 94.14%. This work supports the broader development of VidLLMs by contributing to (1) the creation of datasets for video quality evaluation, (2) interpretable artifact-based analysis beyond language metrics, and (3) the identification of visual risks relevant to factuality and safety.

The Illusion of Role Separation: Hidden Shortcuts in LLM Role Learning (and How to Fix Them)

Large language models (LLMs) that integrate multiple input roles (e.g., system instructions, user queries, external tool outputs) are increasingly prevalent in practice. Ensuring that the model accurately distinguishes messages from each role -- a concept we call \emph{role separation} -- is crucial for consistent multi-role behavior. Although recent work often targets state-of-the-art prompt injection defenses, it remains unclear whether such methods truly teach LLMs to differentiate roles or merely memorize known triggers. In this paper, we examine \emph{role-separation learning}: the process of teaching LLMs to robustly distinguish system and user tokens. Through a \emph{simple, controlled experimental framework}, we find that fine-tuned models often rely on two proxies for role identification: (1) task type exploitation, and (2) proximity to begin-of-text. Although data augmentation can partially mitigate these shortcuts, it generally leads to iterative patching rather than a deeper fix. To address this, we propose reinforcing \emph{invariant signals} that mark role boundaries by adjusting token-wise cues in the model's input encoding. In particular, manipulating position IDs helps the model learn clearer distinctions and reduces reliance on superficial proxies. By focusing on this mechanism-centered perspective, our work illuminates how LLMs can more reliably maintain consistent multi-role behavior without merely memorizing known prompts or triggers.

Emotional Analysis of Fashion Trends Using Social Media and AI: Sentiment Analysis on Twitter for Fashion Trend Forecasting

This study explores the intersection of fashion trends and social media sentiment through computational analysis of Twitter data using the T4SA (Twitter for Sentiment Analysis) dataset. By applying natural language processing and machine learning techniques, we examine how sentiment patterns in fashion-related social media conversations can serve as predictors for emerging fashion trends. Our analysis involves the identification and categorization of fashion-related content, sentiment classification with improved normalization techniques, time series decomposition, statistically validated causal relationship modeling, cross-platform sentiment comparison, and brand-specific sentiment analysis. Results indicate correlations between sentiment patterns and fashion theme popularity, with accessories and streetwear themes showing statistically significant rising trends. The Granger causality analysis establishes sustainability and streetwear as primary trend drivers, showing bidirectional relationships with several other themes. The findings demonstrate that social media sentiment analysis can serve as an effective early indicator of fashion trend trajectories when proper statistical validation is applied. Our improved predictive model achieved 78.35% balanced accuracy in sentiment classification, establishing a reliable foundation for trend prediction across positive, neutral, and negative sentiment categories.

A simple and effective approach for body part recognition on CT scans based on projection estimation

It is well known that machine learning models require a high amount of annotated data to obtain optimal performance. Labelling Computed Tomography (CT) data can be a particularly challenging task due to its volumetric nature and often missing and$/$or incomplete associated meta-data. Even inspecting one CT scan requires additional computer software, or in the case of programming languages $-$ additional programming libraries. This study proposes a simple, yet effective approach based on 2D X-ray-like estimation of 3D CT scans for body region identification. Although body region is commonly associated with the CT scan, it often describes only the focused major body region neglecting other anatomical regions present in the observed CT. In the proposed approach, estimated 2D images were utilized to identify 14 distinct body regions, providing valuable information for constructing a high-quality medical dataset. To evaluate the effectiveness of the proposed method, it was compared against 2.5D, 3D and foundation model (MI2) based approaches. Our approach outperformed the others, where it came on top with statistical significance and F1-Score for the best-performing model EffNet-B0 of 0.980 $\pm$ 0.016 in comparison to the 0.840 $\pm$ 0.114 (2.5D DenseNet-161), 0.854 $\pm$ 0.096 (3D VoxCNN), and 0.852 $\pm$ 0.104 (MI2 foundation model). The utilized dataset comprised three different clinical centers and counted 15,622 CT scans (44,135 labels).

Natural Language Processing tools for Pharmaceutical Manufacturing Information Extraction from Patents

Abundant and diverse data on medicines manufacturing and other lifecycle components has been made easily accessible in the last decades. However, a significant proportion of this information is characterised by not being tabulated and usable for machine learning purposes. Thus, natural language processing tools have been used to build databases in domains such as biomedical and chemical to address this limitation. This has allowed the development of artificial intelligence applications, which have improved drug discovery and treatments. In the pharmaceutical manufacturing context, some initiatives and datasets for primary processing can be found, but the manufacturing of drug products is an area which is still lacking, to the best of our knowledge. This works aims to explore and adapt NLP tools used in other domains to extract information on both primary and secondary manufacturing, employing patents as the main source of data. Thus, two independent, but complementary, models were developed comprising a method to select fragments of text that contain manufacturing data, and a named entity recognition system that enables extracting information on operations, materials, and conditions of a process. For the first model, the identification of relevant sections was achieved using an unsupervised approach combining Latent Dirichlet Allocation and k-Means clustering. The performance of this model measured as a Cohen's kappa between model output and manual revision was higher than 90%. NER model consisted of a deep neural network, and an f1-score micro average of 84.2% was obtained which is comparable to other works. Some considerations for these tools to be used in data extraction are discussed throughout this document.