Computer Science (arXiv)

In this paper, we tackle the problem of performing consistent, unified modifications to a multi-shot video sequence. This task is particularly challenging because multi-shot videos consist of discontinuous temporal segments that vary significantly in viewpoint, camera scale, and subject pose, leading to severe identity drift and cumulative error propagation. Achieving coherent edits requires establishing reliable cross-shot semantic awareness to maintain stable subject appearance and visual continuity across these disjointed boundaries. To address this, we propose MSEditor, the first framework designed specifically for consistent multi-shot video editing. To overcome the scarcity of high-quality multi-shot training data, we repurpose existing multi-view video datasets to provide robust cross-shot supervision. Architecturally, we introduce a Supervisory Adapter that injects this cross-shot information into the diffusion backbone, enabling the model to learn identity-consistent representations. Furthermore, to effectively mitigate cumulative errors and ensure long-range temporal coherence, we design a Cross-Shot Packing strategy that dynamically aggregates information from semantically related shots within the self-attention window. Extensive experiments demonstrate that MSEditor significantly outperforms existing methods on our curated multi-shot video editing benchmark in terms of identity preservation, temporal stability, and overall visual quality.

Large Language Models (LLMs) in real-world applications often face the risks of specially crafted prompts designed to bypass the safety controls. Existing guardrail methods, such as LLM-as-a-judge and cloud-based safety APIs are able to detect unsafe content. However, they often add a delay of about 250-900 ms to each request. This delay is too high for real-time applications, when the system usually needs to respond in less than 100 ms. Furthermore, routing user prompts through external moderation endpoints raises significant data privacy concerns. This paper introduces Reflex-Guard, a lightweight guardrail that runs locally. It uses jailbreak-aware preprocessing, compact sentence-transformer embeddings, and seven fast binary classifiers. Together, these components enable high-accuracy prompt safety filtering with much lower latency than existing solutions. Through systematic evaluation on a strategically balanced dataset of 30,568 samples drawn from five complementary sources, we demonstrate that Reflex-Guard achieves 95.9% recall on harmful prompts at 37.6 ms end-to-end latency. It is faster than existing baselines, including Llama Guard 2 at 255 ms and SafeDecoding at 723 ms. It can detect 100% of GCG suffix attacks and Base64-encoded prompts using the default threshold. However, DrAttack structured prompts required lowering the threshold to 0.03 for optimal detection, as they produced a distinct probability distribution. Reflex-Guard achieves Reflex Efficiency Score (RES) scores up to 16.79, significantly outperforming Llama Guard 2 (11.90) and SafeDecoding (9.80). This analysis offers practical deployment advice and shows that different attack types occupy distinct regions in the embedding probability space.

Cameras are ubiquitous sensors in robotics due to their compact form factor and the perceptual richness captured through visual information. Monocular SLAM enables robots to understand the environment with a minimum setup, however, it inherently suffers from scale ambiguity. A common solution is to provide multi-modal sensor configurations, such as visual-inertial systems, where scale is observable unless the robot navigates under a constant-velocity motion, a common scenario in mobile robotics. With the advent of deep-learning, geometric foundation models have been used to address this problem, but the depths maps are often noisy and scale-inconsistent across frames. In this paper, we propose Scalix, a real-time monocular SLAM framework that achieves metric-scale state estimation by integrating learned depth cues into a probabilistic factor-graph formulation. By augmenting existing monocular depth models with both per-pixel depth uncertainty and per-frame scale uncertainty, Scalix treats scale predictions as independent measurements within its optimization, leading to improved scale consistency through multi-view data associations. Experiments in large-scale outdoor and indoor environments demonstrate state-of-the-art performance on both metric and up-to-scale benchmarks while maintaining real-time operation and generalization.

In Byzantine Agreement (BA), $n$ parties, out of which $t$ can be Byzantine, run a distributed protocol to agree on a common valid input. Traditionally, these protocols have a linear latency and quadratic message complexity, making them impractical at a large scale. In their recent work, Constantinescu, Dufay, Paramonov, and Wattenhofer consider the actual number of byzantine parties $f \leq t$ and work toward decoupling the dependency on $n$ and $t$ in the complexity. They obtain a BA protocol with $\tilde{\mathcal{O}}(n + t\cdot f)$ message complexity and $\tilde{\mathcal{O}}(f)$ round complexity. However, their results are strictly limited to agreement on a binary value. Using the framework given by their work along with novel techniques, we extend these results for BA on an $L$-bit value. With $κ$ being a security parameter, and with optimal resiliency ($t < n/2$ in the synchronous setting or $t < n/3$ otherwise), we obtain: - In synchrony, a deterministic protocol with $\mathcal{O}(n\cdot (L + f \cdot κ))$ bit complexity and $\mathcal{O}(f + \log n)$ round complexity. - In synchrony and partial synchrony, deterministic protocols with $\tilde{\mathcal{O}}(n \cdot κ+ t\cdot (L + f \cdot κ))$ bit complexity and $\mathcal{O}(f)$ round complexity. - In asynchrony, a protocol with $\tilde{\mathcal{O}}(n \cdot κ+ t\cdot(L + t \cdot κ))$ expected bit complexity and expected $\mathcal{O}(1)$ latency.

Academic papers are a primary carrier of scientific knowledge, yet most of this knowledge remains locked in PDFs that are optimized for human reading rather than machine use. For Multimodal Large Language Models (MLLMs), the core challenge is not only perception, but representation: scientific pages interleave text with Structured Academic Elements (SAEs) such as tables, formulas, charts, and pseudocode, whose structure, data, and logic are poorly preserved by common surrogates like Markdown. We therefore propose Compilable Academic Document Parsing (CADP), a paradigm that reconstructs a full page as contextual \LaTeX{} plus executable Python, so that structure-preserving elements and executable chart representations can be reconstructed, recompiled, and directly verified against the source page. To support this setting, we introduce CADP-Bench, an expert-verified benchmark of full academic pages containing tightly coupled text and multiple SAE types, evaluated through a re-injection compilation protocol. We further study current capabilities using SOTA MLLMs and an exploratory multi-agent baseline that incorporates common agentic techniques. Results show that even frontier models still struggle to produce high-fidelity executable reconstructions, highlighting substantial room for improvement in structure-aware scientific document parsing. CADP-Bench is released for future research.

Testing RESTful APIs requires generating sequences of API calls that satisfy dependencies among operations, parameters, and runtime-created resources. Recent LLM-based approaches infer such dependencies and generate test sequences from OpenAPI specifications, but they often treat LLM-inferred relationships as correct without execution-based validation. This can introduce spurious dependencies, miss feasible operation chains, and produce infeasible tests. In this paper, we propose APIPilot}, an execution-validated framework for REST API testing. APIPilot first derives candidate producer-consumer dependencies from OpenAPI specifications using structural heuristics and LLM-based semantic reasoning. It then treats these dependencies as hypotheses and validates them through concrete API executions before using them for test generation. The validated dependencies are organized into a dependency graph from which APIPilot constructs coverage-aware workflows via bounded top-k graph traversal, separating semantic dependency inference from sequence construction. To improve subsequent tests, APIPilot further performs response-driven refinement: runtime responses are analyzed to update resource pools, adjust input-generation constraints, and prune or revise invalid dependency mappings. Empirical evaluation on 16 real-world REST API services shows that APIPilot achieves 92.3% operation coverage, up to 58.6% code coverage, and an 88.1% workflow execution success rate, outperforming both LLM-based and traditional REST API testing baselines. APIPilot also detects 197 unique 5xx failures and specification-execution mismatches, demonstrating the benefit of grounding dependency inference in execution feedback.

Joint-Embedding Predictive Architectures (JEPAs) learn world models by predicting future embeddings, but the objective admits a trivial solution of a constant encoder, so every practical system adds an anti-collapse mechanism (LeCun, 2022; Assran et al., 2023; Bardes et al., 2022; 2024). LeWorldModel (LeWM) prevents collapse with SIGReg, a regularizer that forces the latent distribution to match an isotropic Gaussian: the representation is stabilized by prescribing what it must look like, independently of the environment it models. We argue that the anti-collapse pressure can instead come from the transition data itself. Action-Contrastive Masked Transition Modeling (AC-MTM) keeps LeWM's forward latent-prediction objective and adds a training-only inverse-dynamics head trained with Action-NCE: each latent transition must identify the action that produced it among the other actions in the batch, a discrimination task that a collapsed encoder provably fails. The inverse branch is discarded after training, leaving test-time encoding, forward prediction, planning, and compute identical to LeWM. On four standard pixel-control tasks under a matched planning protocol, AC-MTM trains stably from scratch and matches SIGReg on average. On the harder multi-object OGBench Visual Scene task, results are consistent with the prescribed geometry becoming a bottleneck: AC-MTM reaches 80.0$\pm$2.0% success versus 58.0$\pm$2.0% for SIGReg, improving by 20-24 points in each training seed. A single 50-episode random-policy run gives a 52% baseline estimate. Contrastive inverse dynamics thus provides a distribution-free anti-collapse signal that requires no target network, stop-gradient, pretrained encoder, or reconstruction objective, and we characterize the action-space and observability assumptions under which it holds. We make our code available at https://github.com/jackboyla/action-contrastive-jepa

Ultra-cheap microchips (<$1) are so abundant they've become a 'smart material' integrated and disposable in everyday things. Hidden in our everyday products, we have entirely lost sight of them, yet they account for the vast majority of the >400 billion pieces sold each year. As new technology nodes are released, older ones (from as far back as the 1980s) continue to be produced. These microchips do not exist on their own; they are packaged into every possible item to bring 'smartness', necessary or not; this simultaneously increases their obsolescence. While the latest ICs power our data centres and AI revolution that draws our attention, what about technology so disposable that it has become entirely invisible? We report on our workshop at ICT4S exploring these devices' true costs, and pose challenges to the LOCO community to push back on this system, and develop the skills necessary to create lasting technology and avoid further e-Waste.

This work addresses the orchestration of large-scale Quantum Key Distribution Networks (QKDNs) using Software Defined Networking (SDN). Building on ETSI and ITU specifications, common best practices and architectures are outlined. The main task of the SDN Controller is to aggregate technical key performance indicators (KPI) from the network and, based on these, select the optimal path. Multiple path selection algorithms, based on Dijkstra or a maximum-minimum capacity algorithm, with built-in load balancing are presented. The algorithms were tested in simulations and their performances, and tradeoffs, are discussed. Additional critical aspects related to SDN controlled QKDNs are discussed, such as query batching, multi-path selection and group key capabilities. An oblivious multi-party protocol is proposed for relay path selection in a multi-domain scenario, so providers don't have to disclose sensitive information about their QKDN. These contributions aim to enhance scalability, resilience and interoperability in quantum-secure network infrastructures.

Telegram, with over 450 million daily active users, has introduced Mini Apps---web-based applications running directly within its client. However, this integration introduces notable security risks. As we demonstrate, many Mini Apps store authentication materials---such as session tokens and wallet mnemonic phrases---in plaintext on client devices, exposing users to unauthorized access, impersonation, and financial exploitation. While insecure client-side storage is a known risk in web applications, the Telegram Mini App ecosystem presents a uniquely dangerous combination of factors absent from prior work: no platform-level security review, no storage access restrictions, a financially motivated user base handling live cryptocurrency assets, and a WebView environment that offers weaker protections than standalone browsers. To investigate this threat, we present TENET, a purpose-built auditing tool whose design decisions---pattern selection, entropy thresholds, and charset validation---are grounded in the structural properties of the secrets targeted and empirically validated against a ground-truth dataset. We screened 61 Mini Apps using a stratified, popularity-weighted sampling strategy based on popularity. Of the 37 applications that met our processing criteria and were analyzed, 30 exhibited security flaws, which we classify into three severity tiers: plaintext storage, recoverable encryption, and replayable tokens. Notably, even Telegram's official Wallet exhibits a severe vulnerability that may lead to full account compromise. Following our responsible disclosure, Telegram implemented two new secure-storage APIs, and our post-remediation verification confirmed that its official Wallet no longer exposes the recovery mnemonic in plaintext. Finally, we propose mitigation measures and best practices for both Telegram platform developers and third-party Mini App creators.

Legal consultation questions exhibit multi-level complexity. A single retrieval strategy often leads to over-reasoning for simple questions and poor interpretability for complex ones, making it difficult to meet the requirements for both answer quality and efficiency in high-risk scenarios. To address this issue, this paper proposes CoAL-RAG, a complexity-aware legal retrieval-augmented generation method, which constructs a multi-dimensional evaluation mechanism based on ``question essence'' and ``retrieval consistency'' to enable adaptive routing of retrieval strategies. First, the reasoning demand is quantified according to the logical structure of the question. Then, the discrepancy between semantic retrieval and keyword retrieval is utilized to indirectly reflect problem complexity, thereby selecting the most appropriate retrieval strategy and dynamically filtering contextual information. Experimental results demonstrate that the proposed method significantly outperforms baseline models not only on Chinese legal benchmarks (SocialLawQA, LawBench) but also demonstrates strong cross-jurisdictional generalization on English datasets (LexGLUE, CaseHold). Specifically, on Chinese datasets, the BLEU score improves by 42.5\% and ROUGE-L reaches 3.6 times that of knowledge graph-based methods. On English benchmarks, CoAL-RAG maintains highly competitive accuracy, achieving an optimal balance between generation quality, deep logical reasoning, and system efficiency across different legal systems.

Simultaneously reconstructing and understanding 3D environments is essential for embodied agents. Toward this goal, feed-forward semantic 3D Gaussian Splatting (3DGS) efficiently constructs semantic scene representations from sparse multi-view observations. However, existing methods lack explicit instance discrimination and mainly support category- or phrase-based semantic queries. To this end, we propose GroupForward, an instance-grouped feed-forward Gaussian splatting model that reconstructs geometry, appearance, instance structure, and semantics from sparse, unposed, and uncalibrated multi-view images. Unlike existing methods that attach high-dimensional semantic features to each Gaussian, GroupForward learns compact instance embeddings that group Gaussians into cross-view consistent 3D instances, reformulating feed-forward semantic 3DGS from per-Gaussian semantic feature rendering to instance-level semantic aggregation and propagation. Building on these instance groups, we further propose a Referential Scene Reasoning Framework (RSRF) for complex 3D referring segmentation. RSRF constructs an instance-grouped 3D scene graph and retrieves candidate instances for a given referring expression. A vision-language model then reasons over structured instance evidence and multi-view observations to identify the referred instance among the candidates. RSRF thereby extends language interaction from simple semantic querying to complex referential scene reasoning. Experiments on semantic reconstruction and referential reasoning demonstrate the effectiveness of our instance-grouped reconstruction and reasoning framework.

Large language models increasingly serve as persistent conversational assistants, requiring memory that preserves relevant experience and maintains continuity across interactions. Existing methods improve access to conversational history through long-context processing, selective retrieval, and structured memory organization. However, most systems treat memory access as retrieving relevant past information without first determining which prior interaction state the current turn resumes. This limitation becomes particularly important when conversations interleave multiple tasks, people, and plans that may be interrupted and later revisited. We introduce ArborMem, an online memory framework that represents a long-running conversation as a navigable forest of interaction states. Each branch preserves a locally coherent trajectory, while the forest maintains multiple trajectories that may later be resumed. For each new input, ArborMem localizes the relevant state, restores its branch-local context, and augments it with reusable evidence retrieved across branches, preserving interaction continuity without conflating semantically related but structurally distinct trajectories. Existing long-term memory benchmarks cover diverse memory and reasoning capabilities but do not explicitly isolate branch-structured challenges. We therefore introduce BranchMemEval, a controlled diagnostic benchmark for interleaved and resumable interaction trajectories. Experiments on LongMemEval, LoCoMo, BEAM 100K, and BranchMemEval show that ArborMem outperforms the strongest baselines by 3.36 to 10.31 percentage points on the three established benchmarks and by 5.0 points on BranchMemEval. Its advantage grows under constrained read budgets, while complete memory queries remain below half a second.

Cross-chain bridges, instant cryptocurrency exchanges, and centralized cross-ledger platforms move assets across an increasingly multi-chain ecosystem. However, these systems have repeatedly become targets of high-value attacks and channels for cross-chain money laundering. Cross-chain transactions are substantially harder to analyze than single-chain transactions: no single ledger records an entire cross-chain transfer, its evidence is scattered across the source chain, the destination chain, and off-chain systems, and the availability and reliability of that evidence vary widely across systems. In this paper, we present a systematization of knowledge (SoK) on cross-chain transaction identification and matching. First, we classify deposit and withdrawal identification methods into four approaches and transaction matching methods into three mechanisms: deterministic identifier matching, field-constraint heuristics, and model-assisted matching. We find that their applicability and reported performance are shaped mainly by the evidence the underlying system exposes, and we further examine how matched pairs support downstream attack detection and fund tracing. Second, we assess the availability of existing datasets and artifacts, finding that fewer than half remain obtainable, and distill three artifact failure modes. Finally, we outline four open challenges toward auditable, reproducible, and actionable cross-chain analysis.

Continual pre-training of large language models must acquire new information without erasing old knowledge. Existing replay methods often choose a global old/new mixture and sample uniformly, ignoring that examples differ in how quickly they are forgotten. We formulate continual pre-training as adaptive review scheduling: the training loop should decide not only how much history to replay, but which examples should return at each step. We introduce Spaced Repetition Training (SRT), a continual learning framework inspired by cognitive science, which schedules sample-rehearsal using the SuperMemo-2 (SM-2) algorithm. SRT maintains per-example review state, maps per-example perplexity to a recall-quality signal, and schedules historical examples for retention and new examples for consolidation while leaving the model, objective, and optimizer unchanged. On temporally separated Wikipedia and code corpora, SRT improves the stability-plasticity trade-off, recovering 5 to 37 percentage points of old-knowledge accuracy lost by naive continual pre-training across model scales while preserving or improving new-knowledge acquisition. At larger scale, SRT preserves broad benchmark performance that naive continual pre-training and uniform replay substantially degrade. Experiments with vision and tabular data further suggest that the scheduling principle extends beyond language when paired with an appropriate recall signal.

We present CryptDough, a unified analytics engine for secure multiparty computation (MPC). CryptDough enables multiple distrusting parties to jointly execute a data analysis pipeline on their private inputs and learn nothing beyond the result (e.g., aggregate statistics). Unlike existing MPC solutions that support a single threat model or workload type, CryptDough provides built-in support for cross-domain analytics (relational, time series, ML inference) under various threat models, all within the same system runtime. CryptDough contributes (i) a hierarchical system design that facilitates modularity and extensibility through progressive lowering of abstractions, and (ii) the concept of virtual vectors that enable users to write single-threaded code across all layers of the software stack, while pushing the complexity of communication, parallelization, and memory management down to the execution engine. We show that CryptDough generalizes the functionality of state-of-the-art MPC systems and remains competitive on the analytics they support, often outperforming them by more than $2\times$.

Modern agents operate inside agent harnesses that manage tools, context, and control flow, making the harness a critical part of the agent system. Our original Agent Lightning introduced a disaggregated architecture that connects arbitrary agents to RL training through an LLM endpoint proxy, an approach later adopted by frameworks such as verl Uni-Agent, AReaL 2.0, slime, and Polar. We refer to this paradigm as harnessed agentic RL, where the deploy-time harness directly participates in model post-training. Harnessed agentic RL differs fundamentally from traditional agentic RL: the harness, rather than the training engine, owns the environment interaction loop, while the trainer observes only sequences of LLM request-response pairs. This introduces challenges in retokenization, sample merging, advantage calculation, loss normalization, and backend scheduling, which can substantially affect training stability and effectiveness. We present Agent Lightning v1.0, a lightweight framework for harnessed agentic RL implemented in approximately 3,500 lines of code. It supports arbitrary agent harnesses and serves as a practical testbed for studying these challenges. We evaluate it on instruction-following, search, and coding agents, and provide a complete reproducible pipeline for coding-agent RL. Using only 6K training examples and modest compute, RL improves Qwen3.5-9B on SWE-bench Verified from 41.8% to 56.4%, a 14.6-point absolute gain. We release the complete workflow and training scripts to facilitate reproducible research on harnessed agentic RL.

This preliminary paper outlines a planned evaluation benchmark for Explainable Reinforcement Learning (XRL) methods. Current evaluations rely on functionally-grounded metrics like faithfulness and compactness, and on human-grounded proxies like subjective ratings or prediction accuracy. We suggest evaluating XRL methods by how effectively their generated explanations help to diagnose and fix malfunctioning reinforcement learning (RL) agents. We propose EvalXRL, a benchmark in which a Large Language Model (LLM) coding agent uses different XRL methods to diagnose a held-out malfunction in an RL agent, and then repair it. Our proposed benchmark iterates across (environment $\times$ malfunction $\times$ XRL method) tuples and uses the reward signal of the RL agents to form a final score for each XRL method. The coding agent may use the method interactively: invoke the XRL method, process its output, form new hypotheses on what is broken, and invoke the method again with parameters adjusted for testing these hypotheses. This closed-loop structure may be described as a simplified version of the scientific method. Some XRL methods provide self-evaluations that follow this pattern; we propose the first head-to-head comparison of multiple XRL methods in closed-loop usage.

Scale-up fabrics connecting GPUs and AI accelerators carry tensor transfers together with remote reads, writes, atomics, and notifications over shared target-side receiver resources. Byte-denominated credits protect link buffers and streaming HBM traffic, but poorly represent small operations dominated by Atomic execution or response injection. This paper presents SemaCredit, a receiver controller that admits each remote-memory operation against a vector of target-resource demands and returns each component when its corresponding HBM, Atomic, or response stage completes. In a deterministic event simulator with multipath queues, eight HBM partitions, a serialized Atomic engine, and a response engine, SemaCredit matches a strong per-resource byte baseline on HBM-hotspot traffic while reducing small-operation P99 latency by 52.4% under Atomic contention and 10.2% under response incast. Application-shaped mixes show 57.7% and 14.5% P99 latency improvements for AllReduce-shaped and remote-read-shaped traffic while matching byte credits on HBM-dominated MoE traffic.

Persistent shortages in the surgical workforce and inherent limitations of traditional training methods highlight the necessity of automated, data-driven approaches in surgical education. This study addresses these challenges by introducing a novel, explainable AI-powered framework for automated skill assessment, specifically focusing on cataract surgery. We present the world's largest dataset of cataract surgery videos, comprising 2,000 recordings. Additionally, we propose an AI-powered analytical framework that employs advanced computer vision and signal-processing techniques to automatically evaluate surgical videos to derive objective, quantitative performance indicators that complement or potentially replace subjective scoring methods. A significant advantage of our framework over previous methods lies precisely in its explainability of outputs, elevating it beyond merely an opaque skill classification tool. Through experimental analysis of 83 cataract surgery videos, we demonstrate that the automatically computed metrics exhibit strong correlations with expert-based subjective evaluations, achieving up to 87% accuracy in surgical skill assessment. Each metric was individually examined, and expert surgeons provided subjective ratings using the newly introduced Capsulorhexis Skill Assessment System (CSAS). These subjective assessments were compared with ten objective motion-based metrics extracted through our framework. The results indicated a robust correlation between subjective ratings and automated indicators, underscoring the framework's capacity to accurately model surgical expertise.

We introduce BrainNorm, a normative foundation model, trained and tested on ~66,000 T1-weighted structural MRI (T1w sMRI) scans. By leveraging language-image style contrastive pretraining on healthy cohorts across ages, BrainNorm learns a Semantic Atlas Latent space (SAL), where each scan is represented as a set of atlas-parcel embeddings. This yields parcel-specific healthy aging template trajectories that support age-consistent template matching and localized deviation scoring relative to a subject's chronological age. Across 6 downstream cohorts, BrainNorm demonstrates generalization evaluated across 25 task-setting combinations spanning age estimation, brain-age gap estimation, parcel identification, and single- & multi-disease classification tasks under direct inference, zero-shot, few-shot & full-data linear-probe settings. The resulting deviation patterns in SAL space enable zero-shot tasks for disease prediction using parcel-wise abnormalities. Fine-tuning on healthy-only cohorts of downstream datasets further improves the performance of various tasks. Across all classification tasks, linear probing on BrainNorm's frozen embeddings outperforms 9 baselines finetuned under end-to-end supervision. Furthermore, the localized deviations identified by BrainNorm across various neurodegenerative disorders closely align with established neurodegeneration pathology in clinical literature.

Efforts to accelerate AI and robotics adoption require evidence about where communities are ready to act and where support is still needed. Yet averages across stakeholder groups can obscure relationships that emerge when the same person evaluates different challenges. We analyse a repeated card-based survey in which 982 participants provided 15,200 evaluations of 17 AI and robotics challenges. Each challenge was rated on 1-5 measures of significance, complexity and readiness, where readiness refers to perceived community preparedness and available resources rather than personal competence or realised adoption. Because participants evaluated multiple challenges, the design separates stable between-person differences from challenge-specific within-person deviations. Within the same respondent, a challenge rated one point more complex than usual is associated with about 0.21 points lower readiness (p less than 0.001). By contrast, respondents who generally rate challenges as more complex do not systematically report lower readiness (p=0.29). Significance is positively associated with readiness, while unusually high complexity modestly weakens this alignment. These relationships vary across challenge families, and professional background remains associated with adjusted preparedness. On applied cards, confidence, trust and related perceptions add substantial information about readiness, including for held-out participants. For policymakers and organisations, averaging across stakeholders can hide challenge-specific barriers. Readiness assessments should preserve both differences between stakeholder groups and variation within the same stakeholders across challenges. Effective adoption and literacy strategies should ask not only who appears ready, but which challenges they find unusually difficult and whether the likely constraint concerns implementation, capability, assurance or resources.

Vision-based surgical skill assessment has shown strong in-domain results, yet a fundamental question remains unasked: do these models learn transferable representations of surgical proficiency, or do they merely encode dataset-specific visual patterns? This paper systematically analyzes what limits cross-domain skill transfer between the GOALS and OSATS assessment scales using the LASANA and JIGSAWS datasets. Each evaluated method serves a targeted diagnostic purpose: end-to-end training to test whether supervised skill learning transfers directly, Adaptive Sharpness-Aware Minimization (ASAM) to probe whether flatter loss landscapes improve generalization, and augmentation-based self-supervised and contrastive learning to assess whether domain-invariant pretraining decouples skill from visual context. Transfer is evaluated in both directions using a disjoint-participant held-out test set for JIGSAWS. Results reveal an asymmetry: backbones pretrained on JIGSAWS achieve CCC values of 0.77 to 0.80 on LASANA, closely matching the end-to-end baseline, showing cross-rubric transfer is feasible when the target domain provides consistent supervision. Transfer to JIGSAWS fails across all methods, likely due to annotation inconsistencies. Control experiments with a Kinetics-pretrained backbone suggest task-specific heads carry the majority of the skill prediction burden, while the backbone need only provide adequate spatiotemporal features. These findings offer a new perspective on vision-based skill assessment: the central question of whether skill representations transfer across scoring systems has not been previously investigated. Results indicate the visual component is dominant but not solely responsible for skill prediction; further work is needed to conclusively disentangle transferable skill features from those bound to a specific visual domain.

Fine-grained recognition often involves hierarchical label spaces, where a model may be confident about a coarse semantic concept while remaining uncertain among its descendant classes. Such structured ambiguity requires uncertainty representations that capture both fine-grained classes and intermediate concepts. However, existing tools each capture only half of it: flat evidential classifiers quantify total ignorance with a single vacuity on the leaf frame, and hierarchical classifiers propagate point probabilities with no notion of evidence. Hyper-opinions would unify the two, but their general form is exponential in the label count, and existing hyper-evidential networks either require composite labels to be supplied in the training data or read them off an unstructured weight pattern, with no principled notion of which composites deserve mass. We observe that the taxonomy itself is the missing hyperdomain. Its subtrees and leaf singletons form a linear-size focal family, and one local Dirichlet opinion per branching node induces every composite mass in closed form. The resulting model, H$^2$EDL, can be interpreted in two complementary ways using the same set of parameters. From a prediction perspective, it functions as a hierarchical classifier that preserves consistency across different levels of the label tree. From a probabilistic perspective, it defines a valid tree-structured hyper-opinion, where the mass assigned to each node represents the belief that reaches that node but does not provide sufficient confidence to further specialize into its descendants. On FGVC-Aircraft and DERM12345, H$^2$EDL reduces calibration error by approximately half compared with cross-entropy baselines, with the improvement becoming more pronounced at deeper hierarchy levels and under larger training budgets.

Gender bias or other social biases in large language models (LLMs) are frequently evaluated with question answering or survey benchmarks where the LLM needs to give a response in a predefined answer format. It is well known in survey science that the answer format has a substantial impact on answers, just as LLMs are sensitive to the prompt wording. However, to our knowledge it has not been studied yet how changes in answer format impact the measurement of gender bias in LLMs and their alignment with human response distributions. We evaluate three instruction-tuned models on the BBQ benchmark and OpinionQA survey data across closed-ended, Likert-scaled and open-ended formats, comparing bias measurement and distributional alignment under otherwise identical conditions. We find that answer format does substantially alter measured outcomes, including reversals in order rankings. These differences arise because each format elicits distinct response behaviours, such as forced-choice selection, scale-based distributions and refusal in free-text generation. Our findings highlight the importance of treating answer format as a substantive component of LLM evaluation and motivate multi-format designs for more robust model assessment.