Recent advances in multi-modal generative models have enabled significant progress in instruction-based image editing. However, while these models produce visually plausible outputs, their capacity for knowledge-based reasoning editing tasks remains under-explored. In this paper, we introduce KRIS-Bench (Knowledge-based Reasoning in Image-editing Systems Benchmark), a diagnostic benchmark designed to assess models through a cognitively informed lens. Drawing from educational theory, KRIS-Bench categorizes editing tasks across three foundational knowledge types: Factual, Conceptual, and Procedural. Based on this taxonomy, we design 22 representative tasks spanning 7 reasoning dimensions and release 1,267 high-quality annotated editing instances. To support fine-grained evaluation, we propose a comprehensive protocol that incorporates a novel Knowledge Plausibility metric, enhanced by knowledge hints and calibrated through human studies. Empirical results on 10 state-of-the-art models reveal significant gaps in reasoning performance, highlighting the need for knowledge-centric benchmarks to advance the development of intelligent image editing systems.
Document Visual Question Answering (DocVQA) faces dual challenges in processing lengthy multimodal documents (text, images, tables) and performing cross-modal reasoning. Current document retrieval-augmented generation (DocRAG) methods remain limited by their text-centric approaches, frequently missing critical visual information. The field also lacks robust benchmarks for assessing multimodal evidence selection and integration. We introduce MMDocRAG, a comprehensive benchmark featuring 4,055 expert-annotated QA pairs with multi-page, cross-modal evidence chains. Our framework introduces innovative metrics for evaluating multimodal quote selection and enables answers that interleave text with relevant visual elements. Through large-scale experiments with 60 VLM/LLM models and 14 retrieval systems, we identify persistent challenges in multimodal evidence retrieval, selection, and integration.Key findings reveal advanced proprietary LVMs show superior performance than open-sourced alternatives. Also, they show moderate advantages using multimodal inputs over text-only inputs, while open-source alternatives show significant performance degradation. Notably, fine-tuned LLMs achieve substantial improvements when using detailed image descriptions. MMDocRAG establishes a rigorous testing ground and provides actionable insights for developing more robust multimodal DocVQA systems. Our benchmark and code are available at https://mmdocrag.github.io/MMDocRAG/.
Recent advances in Multi-Modal Large Language Models (MLLMs) have enabled unified processing of language, vision, and structured inputs, opening the door to complex tasks such as logical deduction, spatial reasoning, and scientific analysis. Despite their promise, the reasoning capabilities of MLLMs, particularly those augmented with intermediate thinking traces (MLLMs-T), remain poorly understood and lack standardized evaluation benchmarks. Existing work focuses primarily on perception or final answer correctness, offering limited insight into how models reason or fail across modalities. To address this gap, we introduce the MMMR, a new benchmark designed to rigorously evaluate multi-modal reasoning with explicit thinking. The MMMR comprises 1) a high-difficulty dataset of 1,083 questions spanning six diverse reasoning types with symbolic depth and multi-hop demands and 2) a modular Reasoning Trace Evaluation Pipeline (RTEP) for assessing reasoning quality beyond accuracy through metrics like relevance, consistency, and structured error annotations. Empirical results show that MLLMs-T overall outperform non-thinking counterparts, but even top models like Claude-3.7-Sonnet and Gemini-2.5 Pro suffer from reasoning pathologies such as inconsistency and overthinking. This benchmark reveals persistent gaps between accuracy and reasoning quality and provides an actionable evaluation pipeline for future model development. Overall, the MMMR offers a scalable foundation for evaluating, comparing, and improving the next generation of multi-modal reasoning systems.
Validating scientific hypotheses is a central challenge in biomedical research, and remains difficult for artificial intelligence (AI) agents due to the complexity of real-world data analysis and evidence interpretation. In this work, we present BioDSA-1K, a benchmark designed to evaluate AI agents on realistic, data-driven biomedical hypothesis validation tasks. BioDSA-1K consists of 1,029 hypothesis-centric tasks paired with 1,177 analysis plans, curated from over 300 published biomedical studies to reflect the structure and reasoning found in authentic research workflows. Each task includes a structured hypothesis derived from the original study's conclusions, expressed in the affirmative to reflect the language of scientific reporting, and one or more pieces of supporting evidence grounded in empirical data tables. While these hypotheses mirror published claims, they remain testable using standard statistical or machine learning methods. The benchmark enables evaluation along four axes: (1) hypothesis decision accuracy, (2) alignment between evidence and conclusion, (3) correctness of the reasoning process, and (4) executability of the AI-generated analysis code. Importantly, BioDSA-1K includes non-verifiable hypotheses: cases where the available data are insufficient to support or refute a claim, reflecting a common yet underexplored scenario in real-world science. We propose BioDSA-1K as a foundation for building and evaluating generalizable, trustworthy AI agents for biomedical discovery.
As of 2025, Generative Artificial Intelligence (GenAI) has become a central tool for productivity across industries. Beyond text generation, GenAI now plays a critical role in coding, data analysis, and research workflows. As large language models (LLMs) continue to evolve, it is essential to assess the reliability and accuracy of their outputs, especially in specialized, high-stakes domains like finance. Most modern LLMs transform text into numerical vectors, which are used in operations such as cosine similarity searches to generate responses. However, this abstraction process can lead to misinterpretation of emotional tone, particularly in nuanced financial contexts. While LLMs generally excel at identifying sentiment in everyday language, these models often struggle with the nuanced, strategically ambiguous language found in earnings call transcripts. Financial disclosures frequently embed sentiment in hedged statements, forward-looking language, and industry-specific jargon, making it difficult even for human analysts to interpret consistently, let alone AI models. This paper presents findings from the Santa Clara Microsoft Practicum Project, led by Professor Charlie Goldenberg, which benchmarks the performance of Microsoft's Copilot, OpenAI's ChatGPT, Google's Gemini, and traditional machine learning models for sentiment analysis of financial text. Using Microsoft earnings call transcripts, the analysis assesses how well LLM-derived sentiment correlates with market sentiment and stock movements and evaluates the accuracy of model outputs. Prompt engineering techniques are also examined to improve sentiment analysis results. Visualizations of sentiment consistency are developed to evaluate alignment between tone and stock performance, with sentiment trends analyzed across Microsoft's lines of business to determine which segments exert the greatest influence.
Foundation models leveraging vision-language pretraining have shown promise in chest X-ray (CXR) interpretation, yet their real-world performance across diverse populations and diagnostic tasks remains insufficiently evaluated. This study benchmarks the diagnostic performance and generalizability of foundation models versus traditional convolutional neural networks (CNNs) on multinational CXR datasets. We evaluated eight CXR diagnostic models - five vision-language foundation models and three CNN-based architectures - across 37 standardized classification tasks using six public datasets from the USA, Spain, India, and Vietnam, and three private datasets from hospitals in China. Performance was assessed using AUROC, AUPRC, and other metrics across both shared and dataset-specific tasks. Foundation models outperformed CNNs in both accuracy and task coverage. MAVL, a model incorporating knowledge-enhanced prompts and structured supervision, achieved the highest performance on public (mean AUROC: 0.82; AUPRC: 0.32) and private (mean AUROC: 0.95; AUPRC: 0.89) datasets, ranking first in 14 of 37 public and 3 of 4 private tasks. All models showed reduced performance on pediatric cases, with average AUROC dropping from 0.88 +/- 0.18 in adults to 0.57 +/- 0.29 in children (p = 0.0202). These findings highlight the value of structured supervision and prompt design in radiologic AI and suggest future directions including geographic expansion and ensemble modeling for clinical deployment. Code for all evaluated models is available at https://drive.google.com/drive/folders/1B99yMQm7bB4h1sVMIBja0RfUu8gLktCE
Hyperspectral pansharpening has received much attention in recent years due to technological and methodological advances that open the door to new application scenarios. However, research on this topic is only now gaining momentum. The most popular methods are still borrowed from the more mature field of multispectral pansharpening and often overlook the unique challenges posed by hyperspectral data fusion, such as i) the very large number of bands, ii) the overwhelming noise in selected spectral ranges, iii) the significant spectral mismatch between panchromatic and hyperspectral components, iv) a typically high resolution ratio. Imprecise data modeling especially affects spectral fidelity. Even state-of-the-art methods perform well in certain spectral ranges and much worse in others, failing to ensure consistent quality across all bands, with the risk of generating unreliable results. Here, we propose a hyperspectral pansharpening method that explicitly addresses this problem and ensures uniform spectral quality. To this end, a single lightweight neural network is used, with weights that adapt on the fly to each band. During fine-tuning, the spatial loss is turned on and off to ensure a fast convergence of the spectral loss to the desired level, according to a hysteresis-like dynamic. Furthermore, the spatial loss itself is appropriately redefined to account for nonlinear dependencies between panchromatic and spectral bands. Overall, the proposed method is fully unsupervised, with no prior training on external data, flexible, and low-complexity. Experiments on a recently published benchmarking toolbox show that it ensures excellent sharpening quality, competitive with the state-of-the-art, consistently across all bands. The software code and the full set of results are shared online on https://github.com/giu-guarino/rho-PNN.
The rapid advancement and expanding applications of Audio Large Language Models (ALLMs) demand a rigorous understanding of their trustworthiness. However, systematic research on evaluating these models, particularly concerning risks unique to the audio modality, remains largely unexplored. Existing evaluation frameworks primarily focus on the text modality or address only a restricted set of safety dimensions, failing to adequately account for the unique characteristics and application scenarios inherent to the audio modality. We introduce AudioTrust-the first multifaceted trustworthiness evaluation framework and benchmark specifically designed for ALLMs. AudioTrust facilitates assessments across six key dimensions: fairness, hallucination, safety, privacy, robustness, and authentication. To comprehensively evaluate these dimensions, AudioTrust is structured around 18 distinct experimental setups. Its core is a meticulously constructed dataset of over 4,420 audio/text samples, drawn from real-world scenarios (e.g., daily conversations, emergency calls, voice assistant interactions), specifically designed to probe the multifaceted trustworthiness of ALLMs. For assessment, the benchmark carefully designs 9 audio-specific evaluation metrics, and we employ a large-scale automated pipeline for objective and scalable scoring of model outputs. Experimental results reveal the trustworthiness boundaries and limitations of current state-of-the-art open-source and closed-source ALLMs when confronted with various high-risk audio scenarios, offering valuable insights for the secure and trustworthy deployment of future audio models. Our platform and benchmark are available at https://github.com/JusperLee/AudioTrust.
Unmanned Aerial Vehicles (UAVs) are evolving into language-interactive platforms, enabling more intuitive forms of human-drone interaction. While prior works have primarily focused on high-level planning and long-horizon navigation, we shift attention to language-guided fine-grained trajectory control, where UAVs execute short-range, reactive flight behaviors in response to language instructions. We formalize this problem as the Flying-on-a-Word (Flow) task and introduce UAV imitation learning as an effective approach. In this framework, UAVs learn fine-grained control policies by mimicking expert pilot trajectories paired with atomic language instructions. To support this paradigm, we present UAV-Flow, the first real-world benchmark for language-conditioned, fine-grained UAV control. It includes a task formulation, a large-scale dataset collected in diverse environments, a deployable control framework, and a simulation suite for systematic evaluation. Our design enables UAVs to closely imitate the precise, expert-level flight trajectories of human pilots and supports direct deployment without sim-to-real gap. We conduct extensive experiments on UAV-Flow, benchmarking VLN and VLA paradigms. Results show that VLA models are superior to VLN baselines and highlight the critical role of spatial grounding in the fine-grained Flow setting.
Jailbreak attacks pose a serious threat to large language models (LLMs) by bypassing built-in safety mechanisms and leading to harmful outputs. Studying these attacks is crucial for identifying vulnerabilities and improving model security. This paper presents a systematic survey of jailbreak methods from the novel perspective of stealth. We find that existing attacks struggle to simultaneously achieve toxic stealth (concealing toxic content) and linguistic stealth (maintaining linguistic naturalness). Motivated by this, we propose StegoAttack, a fully stealthy jailbreak attack that uses steganography to hide the harmful query within benign, semantically coherent text. The attack then prompts the LLM to extract the hidden query and respond in an encrypted manner. This approach effectively hides malicious intent while preserving naturalness, allowing it to evade both built-in and external safety mechanisms. We evaluate StegoAttack on four safety-aligned LLMs from major providers, benchmarking against eight state-of-the-art methods. StegoAttack achieves an average attack success rate (ASR) of 92.00%, outperforming the strongest baseline by 11.0%. Its ASR drops by less than 1% even under external detection (e.g., Llama Guard). Moreover, it attains the optimal comprehensive scores on stealth detection metrics, demonstrating both high efficacy and exceptional stealth capabilities. The code is available at https://anonymous.4open.science/r/StegoAttack-Jail66
Quantum machine learning (QML) models, like their classical counterparts, are vulnerable to adversarial attacks, hindering their secure deployment. Here, we report the first systematic experimental robustness benchmark for 20-qubit quantum neural network (QNN) classifiers executed on a superconducting processor. Our benchmarking framework features an efficient adversarial attack algorithm designed for QNNs, enabling quantitative characterization of adversarial robustness and robustness bounds. From our analysis, we verify that adversarial training reduces sensitivity to targeted perturbations by regularizing input gradients, significantly enhancing QNN's robustness. Additionally, our analysis reveals that QNNs exhibit superior adversarial robustness compared to classical neural networks, an advantage attributed to inherent quantum noise. Furthermore, the empirical upper bound extracted from our attack experiments shows a minimal deviation ($3 \times 10^{-3}$) from the theoretical lower bound, providing strong experimental confirmation of the attack's effectiveness and the tightness of fidelity-based robustness bounds. This work establishes a critical experimental framework for assessing and improving quantum adversarial robustness, paving the way for secure and reliable QML applications.
The widespread adoption of Language Models (LMs) across industries is driving interest in deploying these services across the computing continuum, from the cloud to the network edge. This shift aims to reduce costs, lower latency, and improve reliability and privacy. Small Language Models (SLMs), enabled by advances in model compression, are central to this shift, offering a path to on-device inference on resource-constrained edge platforms. This work examines the interplay between edge and cloud deployments, starting from detailed benchmarking of SLM capabilities on single edge devices, and extending to distributed edge clusters. We identify scenarios where edge inference offers comparable performance with lower costs, and others where cloud fallback becomes essential due to limits in scalability or model capacity. Rather than proposing a one-size-fits-all solution, we present platform-level comparisons and design insights for building efficient, adaptive LM inference systems across heterogeneous environments.
As large language models (LLMs) are increasingly deployed in high-stakes settings, their ability to refuse ethically sensitive prompts-such as those involving hate speech or illegal activities-has become central to content moderation and responsible AI practices. While refusal responses can be viewed as evidence of ethical alignment and safety-conscious behavior, recent research suggests that users may perceive them negatively. At the same time, automated assessments of model outputs are playing a growing role in both evaluation and training. In particular, LLM-as-a-Judge frameworks-in which one model is used to evaluate the output of another-are now widely adopted to guide benchmarking and fine-tuning. This paper examines whether such model-based evaluators assess refusal responses differently than human users. Drawing on data from Chatbot Arena and judgments from two AI judges (GPT-4o and Llama 3 70B), we compare how different types of refusals are rated. We distinguish ethical refusals, which explicitly cite safety or normative concerns (e.g., "I can't help with that because it may be harmful"), and technical refusals, which reflect system limitations (e.g., "I can't answer because I lack real-time data"). We find that LLM-as-a-Judge systems evaluate ethical refusals significantly more favorably than human users, a divergence not observed for technical refusals. We refer to this divergence as a moderation bias-a systematic tendency for model-based evaluators to reward refusal behaviors more than human users do. This raises broader questions about transparency, value alignment, and the normative assumptions embedded in automated evaluation systems.
The rapid use of large language models (LLMs) has raised critical concerns regarding the factual reliability of their outputs, especially in low-resource languages such as Urdu. Existing automated fact-checking solutions overwhelmingly focus on English, leaving a significant gap for the 200+ million Urdu speakers worldwide. In this work, we introduce UrduFactCheck, the first comprehensive, modular fact-checking framework specifically tailored for Urdu. Our system features a dynamic, multi-strategy evidence retrieval pipeline that combines monolingual and translation-based approaches to address the scarcity of high-quality Urdu evidence. We curate and release two new hand-annotated benchmarks: UrduFactBench for claim verification and UrduFactQA for evaluating LLM factuality. Extensive experiments demonstrate that UrduFactCheck, particularly its translation-augmented variants, consistently outperforms baselines and open-source alternatives on multiple metrics. We further benchmark twelve state-of-the-art (SOTA) LLMs on factual question answering in Urdu, highlighting persistent gaps between proprietary and open-source models. UrduFactCheck's code and datasets are open-sourced and publicly available at https://github.com/mbzuai-nlp/UrduFactCheck.
Recent advances in large language models (LLMs) and multimodal LLMs (MLLMs) have led to strong reasoning ability across a wide range of tasks. However, their ability to perform mathematical reasoning from spoken input remains underexplored. Prior studies on speech modality have mostly focused on factual speech understanding or simple audio reasoning tasks, providing limited insight into logical step-by-step reasoning, such as that required for mathematical problem solving. To address this gap, we introduce Spoken Math Question Answering (Spoken-MQA), a new benchmark designed to evaluate the mathematical reasoning capabilities of speech-based models, including both cascade models (ASR + LLMs) and end-to-end speech LLMs. Spoken-MQA covers a diverse set of math problems, including pure arithmetic, single-step and multi-step contextual reasoning, and knowledge-oriented reasoning problems, all presented in unambiguous natural spoken language. Through extensive experiments, we find that: (1) while some speech LLMs perform competitively on contextual reasoning tasks involving basic arithmetic, they still struggle with direct arithmetic problems; (2) current LLMs exhibit a strong bias toward symbolic mathematical expressions written in LaTex and have difficulty interpreting verbalized mathematical expressions; and (3) mathematical knowledge reasoning abilities are significantly degraded in current speech LLMs.
Recent advancements in large language models (LLMs) have revolutionized natural language processing (NLP) and expanded their applications across diverse domains. However, despite their impressive capabilities, LLMs have been shown to reflect and perpetuate harmful societal biases, including those based on ethnicity, gender, and religion. A critical and underexplored issue is the reinforcement of caste-based biases, particularly towards India's marginalized caste groups such as Dalits and Shudras. In this paper, we address this gap by proposing DECASTE, a novel, multi-dimensional framework designed to detect and assess both implicit and explicit caste biases in LLMs. Our approach evaluates caste fairness across four dimensions: socio-cultural, economic, educational, and political, using a range of customized prompting strategies. By benchmarking several state-of-the-art LLMs, we reveal that these models systematically reinforce caste biases, with significant disparities observed in the treatment of oppressed versus dominant caste groups. For example, bias scores are notably elevated when comparing Dalits and Shudras with dominant caste groups, reflecting societal prejudices that persist in model outputs. These results expose the subtle yet pervasive caste biases in LLMs and emphasize the need for more comprehensive and inclusive bias evaluation methodologies that assess the potential risks of deploying such models in real-world contexts.
Large language models (LLMs) are increasingly envisioned as decision-support tools in clinical practice, yet safe clinical reasoning demands integrating heterogeneous knowledge bases -- trials, primary studies, regulatory documents, and cost data -- under strict accuracy constraints. Existing evaluations often rely on synthetic prompts, reduce the task to single-hop factoid queries, or conflate reasoning with open-ended generation, leaving their real-world utility unclear. To close this gap, we present MedBrowseComp, the first benchmark that systematically tests an agent's ability to reliably retrieve and synthesize multi-hop medical facts from live, domain-specific knowledge bases. MedBrowseComp contains more than 1,000 human-curated questions that mirror clinical scenarios where practitioners must reconcile fragmented or conflicting information to reach an up-to-date conclusion. Applying MedBrowseComp to frontier agentic systems reveals performance shortfalls as low as ten percent, exposing a critical gap between current LLM capabilities and the rigor demanded in clinical settings. MedBrowseComp therefore offers a clear testbed for reliable medical information seeking and sets concrete goals for future model and toolchain upgrades. You can visit our project page at: https://moreirap12.github.io/mbc-browse-app/
Respiratory viral infections pose a global health burden, yet the cellular immune responses driving protection or pathology remain unclear. Natural infection cohorts often lack pre-exposure baseline data and structured temporal sampling. In contrast, inoculation and vaccination trials generate insightful longitudinal transcriptomic data. However, the scattering of these datasets across platforms, along with inconsistent metadata and preprocessing procedure, hinders AI-driven discovery. To address these challenges, we developed the Human Respiratory Viral Immunization LongitudinAl Gene Expression (HR-VILAGE-3K3M) repository: an AI-ready, rigorously curated dataset that integrates 14,136 RNA-seq profiles from 3,178 subjects across 66 studies encompassing over 2.56 million cells. Spanning vaccination, inoculation, and mixed exposures, the dataset includes microarray, bulk RNA-seq, and single-cell RNA-seq from whole blood, PBMCs, and nasal swabs, sourced from GEO, ImmPort, and ArrayExpress. We harmonized subject-level metadata, standardized outcome measures, applied unified preprocessing pipelines with rigorous quality control, and aligned all data to official gene symbols. To demonstrate the utility of HR-VILAGE-3K3M, we performed predictive modeling of vaccine responders and evaluated batch-effect correction methods. Beyond these initial demonstrations, it supports diverse systems immunology applications and benchmarking of feature selection and transfer learning algorithms. Its scale and heterogeneity also make it ideal for pretraining foundation models of the human immune response and for advancing multimodal learning frameworks. As the largest longitudinal transcriptomic resource for human respiratory viral immunization, it provides an accessible platform for reproducible AI-driven research, accelerating systems immunology and vaccine development against emerging viral threats.
This paper introduces a novel surrogate modeling framework for aerodynamic applications based on Neural Fields. The proposed approach, MARIO (Modulated Aerodynamic Resolution Invariant Operator), addresses non parametric geometric variability through an efficient shape encoding mechanism and exploits the discretization-invariant nature of Neural Fields. It enables training on significantly downsampled meshes, while maintaining consistent accuracy during full-resolution inference. These properties allow for efficient modeling of diverse flow conditions, while reducing computational cost and memory requirements compared to traditional CFD solvers and existing surrogate methods. The framework is validated on two complementary datasets that reflect industrial constraints. First, the AirfRANS dataset consists in a two-dimensional airfoil benchmark with non-parametric shape variations. Performance evaluation of MARIO on this case demonstrates an order of magnitude improvement in prediction accuracy over existing methods across velocity, pressure, and turbulent viscosity fields, while accurately capturing boundary layer phenomena and aerodynamic coefficients. Second, the NASA Common Research Model features three-dimensional pressure distributions on a full aircraft surface mesh, with parametric control surface deflections. This configuration confirms MARIO's accuracy and scalability. Benchmarking against state-of-the-art methods demonstrates that Neural Field surrogates can provide rapid and accurate aerodynamic predictions under the computational and data limitations characteristic of industrial applications.
The automatic analysis of document layouts in digital-born PDF documents remains a challenging problem due to the heterogeneous arrangement of textual and nontextual elements and the imprecision of the textual metadata in the Portable Document Format. In this work, we benchmark Graph Neural Network (GNN) architectures for the task of fine-grained layout classification of text blocks from digital native documents. We introduce two graph construction structures: a k-closest-neighbor graph and a fully connected graph, and generate node features via pre-trained text and vision models, thus avoiding manual feature engineering. Three experimental frameworks are evaluated: single-modality (text or visual), concatenated multimodal, and dual-branch multimodal. We evaluated four foundational GNN models and compared them with the baseline. Our experiments are specifically conducted on a rich dataset of public affairs documents that includes more than 20 sources (e.g., regional and national-level official gazettes), 37K PDF documents, with 441K pages in total. Our results demonstrate that GraphSAGE operating on the k-closest-neighbor graph in a dual-branch configuration achieves the highest per-class and overall accuracy, outperforming the baseline in some sources. These findings confirm the importance of local layout relationships and multimodal fusion exploited through GNNs for the analysis of native digital document layouts.