Large Language Models (LLMs) recently achieved great success in medical text summarization by simply using in-context learning. However, these recent efforts do not perform fine-grained evaluations under difficult settings where LLMs might fail. They typically report performance scores over the entire dataset. Through our benchmarking study, we show that LLMs show a significant performance drop for data points with high concentration of out-of-vocabulary (OOV) words or with high novelty. Vocabulary adaptation is an intuitive solution to this vocabulary mismatch issue where the LLM vocabulary gets updated with certain expert domain (here, medical) words or subwords. An interesting finding from our study is that Llama-3.1, even with a vocabulary size of around 128K tokens, still faces over-fragmentation issue with medical words. To that end, we show vocabulary adaptation helps improve the LLM summarization performance even in difficult settings. Through extensive experimentation of multiple vocabulary adaptation strategies, two continual pretraining strategies, and three benchmark medical summarization datasets, we gain valuable insights into the role of vocabulary adaptation strategies for customizing LLMs to the medical domain. We also performed a human evaluation study with medical experts where they found that vocabulary adaptation results in more relevant and faithful summaries. Our codebase is made publicly available at https://github.com/gb-kgp/LLM-MedicalSummarization-Benchmark.

Metal additive manufacturing (AM) involves complex interdependencies among processes, materials, feedstock, and post-processing steps. However, the underlying relationships and domain knowledge remain fragmented across literature and static databases that often require expert-level queries, limiting their applicability in design and planning. To address these limitations, we develop a novel and structured knowledge graph (KG), representing 53 distinct metals and alloys across seven material categories, nine AM processes, four feedstock types, and corresponding post-processing requirements. A large language model (LLM) interface, guided by a few-shot prompting strategy, enables natural language querying without the need for formal query syntax. The system supports a range of tasks, including compatibility evaluation, constraint-based filtering, and design for AM (DfAM) guidance. User queries in natural language are normalized, translated into Cypher, and executed on the KG, with results returned in a structured format. This work introduces the first interactive system that connects a domain-specific metal AM KG with an LLM interface, delivering accessible and explainable decision support for engineers and promoting human-centered tools in manufacturing knowledge systems.

As large language models (LLMs) gain popularity, their vulnerability to adversarial attacks emerges as a primary concern. While fine-tuning models on domain-specific datasets is often employed to improve model performance, it can inadvertently introduce vulnerabilities within the underlying model. In this work, we investigate Accidental Vulnerability, unexpected vulnerabilities arising from characteristics of fine-tuning data. We begin by identifying potential correlation factors such as linguistic features, semantic similarity, and toxicity across multiple experimental datasets. We then evaluate the adversarial robustness of these fine-tuned models, analyzing persona shifts and interpretability traits to understand how dataset factors contribute to attack success rates. Lastly, we explore causal relationships that offer new insights into adversarial defense strategies, highlighting the crucial role of dataset design in preserving model alignment. Our code is available at https://github.com/psyonp/accidental_vulnerability.

A two stage novel generation framework (outline -> section outline -> manuscript) is widely used in long novel generation,(e.g., \textsc{DOME}, \textsc{Plan\&Write}, \textsc{Long Writer}), but study of such framework in ultra long novel(>1M words) reconstruction is little. Building on recent text compression methods (\textsc{LLMZip}, \textsc{LLM2Vec}), we conduct an information-theoretic analysis to quantify semantic distortion under different compression-expansion ratios. We examine how outline length affects information preservation. Experiments on ultra-long novels show that the optimal compression-expansion ratio significantly reduces semantic distortion compared to other non-optimal compression-expansion ratio.

One of the goals of fairness research in NLP is to measure and mitigate stereotypical biases that are propagated by NLP systems. However, such work tends to focus on single axes of bias (most often gender) and the English language. Addressing these limitations, we contribute the first study of multilingual intersecting country and gender biases, with a focus on occupation recommendations generated by large language models. We construct a benchmark of prompts in English, Spanish and German, where we systematically vary country and gender, using 25 countries and four pronoun sets. Then, we evaluate a suite of 5 Llama-based models on this benchmark, finding that LLMs encode significant gender and country biases. Notably, we find that even when models show parity for gender or country individually, intersectional occupational biases based on both country and gender persist. We also show that the prompting language significantly affects bias, and instruction-tuned models consistently demonstrate the lowest and most stable levels of bias. Our findings highlight the need for fairness researchers to use intersectional and multilingual lenses in their work.

Recently, natural language has been the primary medium for human-robot interaction. However, its inherent lack of spatial precision introduces challenges for robotic task definition such as ambiguity and verbosity. Moreover, in some public settings where quiet is required, such as libraries or hospitals, verbal communication with robots is inappropriate. To address these limitations, we introduce the Robotic Visual Instruction (RoVI), a novel paradigm to guide robotic tasks through an object-centric, hand-drawn symbolic representation. RoVI effectively encodes spatial-temporal information into human-interpretable visual instructions through 2D sketches, utilizing arrows, circles, colors, and numbers to direct 3D robotic manipulation. To enable robots to understand RoVI better and generate precise actions based on RoVI, we present Visual Instruction Embodied Workflow (VIEW), a pipeline formulated for RoVI-conditioned policies. This approach leverages Vision-Language Models (VLMs) to interpret RoVI inputs, decode spatial and temporal constraints from 2D pixel space via keypoint extraction, and then transform them into executable 3D action sequences. We additionally curate a specialized dataset of 15K instances to fine-tune small VLMs for edge deployment,enabling them to effectively learn RoVI capabilities. Our approach is rigorously validated across 11 novel tasks in both real and simulated environments, demonstrating significant generalization capability. Notably, VIEW achieves an 87.5% success rate in real-world scenarios involving unseen tasks that feature multi-step actions, with disturbances, and trajectory-following requirements. Project website: https://robotic-visual-instruction.github.io/

T ext-to-video (T2V) generative models have rapidly advanced and found widespread applications across fields like entertainment, education, and marketing. However, the adversarial vulnerabilities of these models remain rarely explored. W e observe that in T2V generation tasks, the generated videos often contain substantial redundant information not explicitly specified in the text prompts, such as environmental elements, secondary objects, and additional details, providing opportunities for malicious attackers to embed hidden harmful content. Exploiting this inherent redundancy, we introduce BadVideo, the first backdoor attack framework tailored for T2V generation. Our attack focuses on designing target adversarial outputs through two key strategies: (1) Spatio-T emporal Composition, which combines different spatiotemporal features to encode malicious information; (2) Dynamic Element Transformation, which introduces transformations in redundant elements over time to convey malicious information. Based on these strategies, the attacker's malicious target seamlessly integrates with the user's textual instructions, providing high stealthiness. Moreover, by exploiting the temporal dimension of videos, our attack successfully evades traditional content moderation systems that primarily analyze spatial information within individual frames. Extensive experiments demonstrate that BadVideo achieves high attack success rates while preserving original semantics and maintaining excellent performance on clean inputs. Overall, our work reveals the adversarial vulnerability of T2V models, calling attention to potential risks and misuse.

Despite increasing AI chatbot deployment in public discourse, empirical evidence on their capacity to foster intercultural empathy remains limited. Through a randomized experiment, we assessed how different AI deliberation approaches--cross-cultural deliberation (presenting other-culture perspectives), own-culture deliberation (representing participants' own culture), and non-deliberative control--affect intercultural empathy across American and Latin American participants. Cross-cultural deliberation increased intercultural empathy among American participants through positive emotional engagement, but produced no such effects for Latin American participants, who perceived AI responses as culturally inauthentic despite explicit prompting to represent their cultural perspectives. Our analysis of participant-driven feedback, where users directly flagged and explained culturally inappropriate AI responses, revealed systematic gaps in AI's representation of Latin American contexts that persist despite sophisticated prompt engineering. These findings demonstrate that current approaches to AI cultural alignment--including linguistic adaptation and explicit cultural prompting--cannot fully address deeper representational asymmetries in AI systems. Our work advances both deliberation theory and AI alignment research by revealing how the same AI system can simultaneously promote intercultural understanding for one cultural group while failing for another, with critical implications for designing equitable AI systems for cross-cultural democratic discourse.

Large Language Models (LLMs) have transformed human-computer interaction by enabling natural language-based communication with AI-powered chatbots. These models are designed to be intuitive and user-friendly, allowing users to articulate requests with minimal effort. However, despite their accessibility, studies reveal that users often struggle with effective prompting, resulting in inefficient responses. Existing research has highlighted both the limitations of LLMs in interpreting vague or poorly structured prompts and the difficulties users face in crafting precise queries. This study investigates learner-AI interactions through an educational experiment in which participants receive structured guidance on effective prompting. We introduce and compare three types of prompting guidelines: a task-specific framework developed through a structured methodology and two baseline approaches. To assess user behavior and prompting efficacy, we analyze a dataset of 642 interactions from 107 users. Using Von NeuMidas, an extended pragmatic annotation schema for LLM interaction analysis, we categorize common prompting errors and identify recurring behavioral patterns. We then evaluate the impact of different guidelines by examining changes in user behavior, adherence to prompting strategies, and the overall quality of AI-generated responses. Our findings provide a deeper understanding of how users engage with LLMs and the role of structured prompting guidance in enhancing AI-assisted communication. By comparing different instructional frameworks, we offer insights into more effective approaches for improving user competency in AI interactions, with implications for AI literacy, chatbot usability, and the design of more responsive AI systems.

Existing Speech Language Model (SLM) scaling analysis paints a bleak picture. It predicts that SLMs require much more compute and data compared to text, leading some to question the feasibility of training high-quality SLMs. However, modern SLMs are often initialised from pre-trained TextLMs using speech-text interleaving to allow knowledge transfer. This raises the question - "Do interleaved SLMs scale more efficiently than textless-SLMs?" In this paper we answer a resounding yes! We conduct scaling analysis of interleaved SLMs by training several dozen and analysing the scaling trends. We see that under this setup SLMs scale more efficiently with compute. Additionally, our results indicate that the scaling dynamics significantly differ from textless-SLMs, suggesting one should allocate notably more of the compute budget to increasing model size over training tokens. We also study the role of synthetic data and TextLM model families in unlocking this potential. Results suggest that our scaled up model achieves comparable semantic speech performance to leading models, while using less compute and data. We open source models, samples, and data - https://pages.cs.huji.ac.il/adiyoss-lab/sims/ .