Research Papers

Grading knee osteoarthritis (KOA) on plain radiographs is poorly reproducible across readers. A single-grade disagreement on the Kellgren-Lawrence (KL) scale can alter surgical management or redirect a patient from conservative therapy to intra-articular injection. Meanwhile, deep learning models that outperform human readers often offer no explanation for their decisions. We present Knee-xRAI, a pipeline that decomposes the grading process by mimicking clinical radiological workflows. It independently measures joint space narrowing (JSN), osteophytes, and subchondral sclerosis, then combines these findings into an explainable KL grade. Specifically, a U-Net++ architecture quantifies JSN via contour segmentation, an SE-ResNet-50 multi-task network grades osteophytes per anatomical site on the OARSI scale, and a hybrid texture-CNN detects binary sclerosis. This pipeline yields a 50-dimensional feature vector evaluated via an XGBoost-SHAP classifier (Path A, audit) and a ConvNeXt hybrid predictor (Path B, deployed). On 8,260 OAI-derived radiographs, the JSN module achieved a Dice score of 0.8909 and an mJSW ICC of 0.8674. Path A reached a QWK of 0.6294 and an AUC of 0.8046, confirming the structured feature vector carries substantial diagnostic signal. Path B achieved a QWK of 0.8436 and an AUC of 0.9017. SHAP analysis identifies JSN as the dominant feature, with osteophytes adding a consistent increment and sclerosis contributing marginally. Removing JSN evidence collapses KL3-KL4 recall while early grades remain intact, aligning with the KL diagnostic criteria. Knee-xRAI grounds every prediction in an auditable chain of measured radiographic findings, providing clinical transparency at the point of care.

Large language models are being deployed as mental health support agents at scale, yet only 16% of LLM-based chatbot interventions have undergone rigorous clinical efficacy testing, and simulations reveal psychological deterioration in over one-third of cases. We evaluate four generative models on 250 Prolonged Exposure (PE) therapy scenarios and 146 CBT cognitive restructuring exercises (plus 29 severity-escalated variants), scored by a three-judge LLM panel. All models scored near-perfectly on surface acknowledgment (~0.91-1.00) while therapeutic appropriateness collapsed to 0.22-0.33 at the highest severity for three of four models, with protocol fidelity reaching zero for two. Under CBT severity escalation, one model's task completeness dropped from 92% to 71% while the frontier model's safety-interference score fell from 0.99 to 0.61. We identify a systematic, modality-spanning failure: RLHF safety alignment disrupts the therapeutic mechanism of action by grounding patients during imaginal exposure, offering false reassurance, inserting crisis resources into controlled exercises, and refusing to challenge distorted cognitions mentioning self-harm in PE; and through task abandonment or safety-preamble insertion during CBT cognitive restructuring. These findings motivate a five-axis evaluation framework (protocol fidelity, hallucination risk, behavioral consistency, crisis safety, demographic robustness), mapped onto FDA SaMD and EU AI Act requirements. We argue that no AI mental health system should proceed to deployment without passing multi-axis evaluation across all five dimensions.

Despite growing concerns about the risks of Generative AI (GenAI), there is limited understanding of public perceptions of these risks and their associated failure modes -- defined as recurring patterns of sociotechnical breakdown across the GenAI lifecycle that contribute to risks of real-world harm. To address this gap, we present a survey instrument, validated with eight subject matter experts and deployed on a sample of 960 U.S.-based participants, to assess awareness and perceptions of GenAI's failure modes, their associated risks, and stakeholder responsibilities to address them. To support realism and content validity, our instrument is structured around scenarios grounded in publicly reported incidents and a taxonomy of GenAI's failure modes. Findings suggest that our instrument is (1) effective for assessing risk awareness and perceptions in a way that is grounded in people's current contexts of use, yet is extensible to new contexts that will inevitably arise; and (2) potentially useful for informing the design of AI literacy tools and interventions. We argue for AI literacy and governance approaches that align with how people encounter and reason about GenAI in everyday life.

Evaluating Post-hoc Explanations of the Transformer-based Genome Language Model DNABERT-2

Isabel Kurth, Paulo Yanez Sarmiento, Bernhard Y. Renard

Explaining deep neural network predictions on genome sequences enables biological insight and hypothesis generation-often of greater interest than predictive performance alone. While explanations of convolutional neural networks (CNNs) have been shown to capture relevant patterns in genome sequences, it is unclear whether this transfers to more expressive Transformer-based genome language models (gLMs). To answer this question, we adapt AttnLRP, an extension of layer-wise relevance propagation to the attention mechanism, and apply it to the state-of-the-art gLM DNABERT-2. Thereby, we propose strategies to transfer explanations from token and nucleotide level. We evaluate the adaption of AttnLRP on genomic datasets using multiple metrics. Further, we provide an extensive comparison between the explanations of DNABERT-2 and a baseline CNN. Our results demonstrate that AttnLRP yields reliable explanations corresponding to known biological patterns. Hence, like CNNs, gLMs can also help derive biological insights. This work contributes to the explainability of gLMs and addresses the comparability of relevance attributions across different architectures.

Interpretability tools are increasingly used to analyze failures of Large Language Models (LLMs), yet prior work largely focuses on short prompts or toy settings, leaving their behavior on commonly used benchmarks underexplored. To address this gap, we study contrastive, LRP-based attribution as a practical tool for analyzing LLM failures in realistic settings. We formulate failure analysis as \textit{contrastive attribution}, attributing the logit difference between an incorrect output token and a correct alternative to input tokens and internal model states, and introduce an efficient extension that enables construction of cross-layer attribution graphs for long-context inputs. Using this framework, we conduct a systematic empirical study across benchmarks, comparing attribution patterns across datasets, model sizes, and training checkpoints. Our results show that this token-level contrastive attribution can yield informative signals in some failure cases, but is not universally applicable, highlighting both its utility and its limitations for realistic LLM failure analysis. Our code is available at: https://aka.ms/Debug-XAI.

Federal agencies are increasingly deploying large language models (LLMs) to process public comments submitted during notice-and-comment rulemaking, the primary mechanism through which citizens influence federal regulation. Whether these systems treat all public input equally remains largely untested. Using a counterfactual design, we held comment content constant and varied only the commenter's demographic attribution -- race, gender, and socioeconomic status -- to test whether eight LLMs available for federal use produce differential summaries of identical comments. We processed 182 public comments across 32 identity conditions, generating over 106,000 summaries. Occupation was the only identity signal to produce consistent differential treatment: the same comment attributed to a street vendor, compared to a financial analyst, received a summary that preserved less of the original meaning, used simpler language, and shifted emotional tone. This pattern held across all names, prompts, models, and regulatory contexts tested. Race effects were inconsistent and appeared driven by specific name tokens rather than racial categories; gender effects were absent. Writing quality predicted summarization outcomes through argument substance rather than surface mechanics; experimentally injected spelling and grammar errors had negligible effects. The magnitude of occupation-based differential treatment varied by model provider, meaning that selecting a model implicitly selects a level of fairness -- a dimension that current procurement frameworks such as FedRAMP do not evaluate. These findings suggest that socioeconomic signals warrant attention in AI fairness assessments for government information systems, and that fairness benchmarks could be incorporated into existing federal IT procurement processes.

RefereeBench: Are Video MLLMs Ready to be Multi-Sport Referees

Yichen Xu, Yuanhang Liu, Chuhan Wang +5

While Multimodal Large Language Models (MLLMs) excel at generic video understanding, their ability to support specialized, rule-grounded decision-making remains insufficiently explored. In this paper, we introduce RefereeBench, the first large-scale benchmark for evaluating MLLMs as automatic sports referees. Spanning 11 sports with 925 curated videos and 6,475 QA pairs, RefereeBench evaluates five core officiating abilities: foul existence, foul and penalty classification, foul and penalty reasoning, entity perception, and temporal grounding. The benchmark is fully human-annotated to ensure high-quality annotations grounded in authentic officiating logic and multimodal evidence. Extensive evaluations of state-of-the-art MLLMs show that even the strongest models, such as Doubao-Seed-1.8 and Gemini-3-Pro, achieve only around 60% accuracy, while the strongest open-source model, Qwen3-VL, reaches only 47%. These results indicate that current models remain far from being reliable sports referees. Further analysis shows that while models can often identify incidents and involved entities, they struggle with rule application and temporal grounding, and frequently over-call fouls on normal clips. Our benchmark highlights the need for future MLLMs that better integrate domain knowledge and multimodal understanding, advancing trustworthy AI-assisted officiating and broader multimodal decision-making.

Towards Rigorous Explainability by Feature Attribution

Olivier Létoffé, Xuanxiang Huang, Joao Marques-Silva

For around a decade, non-symbolic methods have been the option of choice when explaining complex machine learning (ML) models. Unfortunately, such methods lack rigor and can mislead human decision-makers. In high-stakes uses of ML, the lack of rigor is especially problematic. One prime example of provable lack of rigor is the adoption of Shapley values in explainable artificial intelligence (XAI), with the tool SHAP being a ubiquitous example. This paper overviews the ongoing efforts towards using rigorous symbolic methods of XAI as an alternative to non-rigorous non-symbolic approaches, concretely for assigning relative feature importance.

As companies enter the race for agentic AI adoption, fears surface around agentic autonomy and its subsequent risks. These fears compound as companies scale their agentic AI adoption with low-code applications, without a comparable scaling in their governance processes and expertise resulting in a phenomenon known as "Agent Sprawl". While shadow AI tools can help with agentic discovery and identification, few observability tools offer insights into the agents' configuration and settings or the decision-making process during agent-to-agent communication and orchestration. This paper explores AI governance professionals' concerns in enterprise settings, while offering design-time and runtime explainability techniques as suggested by AI governance experts for addressing those fears. Finally, we provide a preliminary prototype of an Agentic AI Card that can help companies feel at ease deploying agents at scale.

This paper advances a methodological proposal for safety research in agentic AI. As systems acquire planning, memory, tool use, persistent identity, and sustained interaction, safety can no longer be analysed primarily at the level of the isolated model. Population-level risks arise from structured interaction among agents, through processes of communication, observation, and mutual influence that shape collective behaviour over time. As the object of analysis shifts, a methodological gap emerges. Approaches focused either on single agents or on aggregate outcomes do not identify the interaction-level mechanisms that generate collective risks or the design variables that control them. A framework is required that links local interaction structure to population-level dynamics in a causally explicit way, allowing both explanation and intervention. We introduce two linked concepts. Agentic microphysics defines the level of analysis: local interaction dynamics where one agent's output becomes another's input under specific protocol conditions. Generative safety defines the methodology: growing phenomena and elicit risks from micro-level conditions to identify sufficient mechanisms, detect thresholds, and design effective interventions.

Mapping the spatial distribution of species is essential for conservation policy and invasive species management. Species distribution models (SDMs) are the primary tools for this task, serving two purposes: achieving robust predictive performance while providing ecological insights into the driving factors of distribution. However, the increasing complexity of deep learning SDMs has made extracting these insights more challenging. To reconcile these objectives, we propose the first implementation of concept-based Explainable AI (XAI) for SDMs. We leverage the Robust TCAV (Testing with Concept Activation Vectors) methodology to quantify the influence of landscape concepts on model predictions. To enable this, we provide a new open-access landscape concept dataset derived from high-resolution multispectral and LiDAR drone imagery. It includes 653 patches across 15 distinct landscape concepts and 1,450 random reference patches, designed to suit a wide range of species. We demonstrate this approach through a case study of two aquatic insects, Plecoptera and Trichoptera, using two Convolutional Neural Networks and one Vision Transformer. Results show that concept-based XAI helps validate SDMs against expert knowledge while uncovering novel associations that generate new ecological hypotheses. Robust TCAV also provides landscape-level information, useful for policy-making and land management. Code and datasets are publicly available.

We present a neurosymbolic approach, i.e. combine symbolic and subsymbolic artificial intelligence, to validating offer documents in regulated public institutions. We employ a language model to extract information and then aggregate it with an LTN (Logic Tensor Network) to make an auditable decision. In regulated public institutions, decisions must be made in a manner that is both factually correct and legally verifiable. Our neurosymbolic approach allows existing domain-specific knowledge to be linked to the semantic text understanding of language models. The decisions resulting from our pipeline can be justified by predicate values, rule truth values, and corresponding text passages. Our experiments on a real corpus show that the proposed pipeline achieves performance comparable to existing models, but its key advantage lies in its interpretability, modular predicate extraction, and explicit support for XAI (Explainable AI).