Papers

On a delivery platform, personalized store ranking greatly influences what users find and order. Unlike digital-only domains, candidate stores are local and bound by real-time availability and delivery operations. One central modeling tension is between surfacing new stores for trial and preserving ranking quality for sessions with reorder intent. We present Universal Venue Ranker (UVR), a production system deployed at Wolt that pairs a bidirectional transformer encoder for sequential user modeling with a GBDT ranker integrating contextual, user, and store features. Trained across all stores and domains of a country while enforcing local delivery constraints at inference, UVR replaces four previously separate ranking models (three for restaurants, one for retail) with a single unified system. Label smoothing and trial-biased sample weighting steer the model toward new stores, lifting offline trial MRR by +12% to +30% over production while regressing reorder MRR in five of six countries. These regressions leave Global CVR, our core online metric, which blends trial and reorder sessions, statistically unchanged. We validate UVR in three consecutive A/B tests, the first two across Wolt's largest operating markets and the third spanning all operating countries and both domains. UVR V1 delivers +5.5% Merchant Trial Rate and +0.16% Global CVR over the previous production ranker; V2 adds a further +0.45% Merchant Trial Rate on top; and V3, our cross-domain unification of the restaurant and retail rankers, adds a further +1.31% Retail Merchant Trial Rate, together accounting for substantial incremental gross order value and a materially simplified serving stack.

Vision-language models (VLMs) can generate routes directly from aerial imagery for off-road navigation, but their predictions provide no indication of reliability. We present UDAV, an Uncertainty-Driven Adaptive VLM Waypoint Planner for UAV-guided UGV navigation. UDAV draws multiple stochastic trajectory predictions, selects their medoid as a self-consistent nominal route, and estimates predictive uncertainty from their spatial dispersion. When the maximum uncertainty across interior waypoints exceeds a threshold, UDAV invokes a reconsideration stage; otherwise, it returns the medoid directly. We evaluate UDAV on 400 held-out trajectory queries from two UAV flights. Stochastic medoid selection reduces the mean average displacement error (ADE) from 147.4 pixels for a deterministic prediction to 115.9 pixels. The complete planner achieves a mean ADE of 110.4 pixels, a 25.1% reduction relative to deterministic planning, while producing valid trajectories for all queries. UDAV also yields the lowest 90th- and 95th-percentile errors among all evaluated configurations, including a higher-budget K=10 consensus baseline. Relative to the K=5 medoid, UDAV reduces these errors from 225.3 and 326.0 pixels to 199.0 and 290.8 pixels, respectively. These results demonstrate that stochastic VLM predictions provide both a stronger nominal route and an actionable uncertainty signal for selectively mitigating large planning errors.

When foundation models describe people, recent work in AI fairness, accessibility, and ethics recommends avoiding inferred identity labels (e.g., "she", "his") in favor of seemingly "objective" physical descriptions (e.g., "short hair", "a defined jawline"). Yet whether such descriptive language achieves gender-neutral communication remains an open empirical question. To study this, we introduce GAPA (Gender Associations of Physical Attributes), a dataset of 316 common physical attributes drawn from diverse sources, paired with 14,706 gender-association ratings from 304 US-based annotators. Results show that physical descriptions carry structured and graded gender associations among readers, with more consistent and distinctive associations for women and men than for non-binary identities. Next, we evaluate 16 LLMs across model families, sizes, and post-training variants against human ratings. The models partially recover human associations but exhibit systematic alignment biases, including compressed rating distributions, weaker alignment for associations with men, and asymmetric abstention that disproportionately targets the non-binary category. Finally, we release the best-performing proxy model trained to predict humans' gender associations of descriptive language and demonstrate its utility through a sociolinguistic analysis of character descriptions in LitBank. Together, our findings provide the first empirical evidence that seemingly "objective" physical descriptions can retain systematic gender associations in human interpretation, and uncover systematic patterns of model-human misalignment. This challenges the assumption that replacing explicit gender labels with physical descriptions necessarily yields gender-neutral communication, and highlights downstream challenges in using such descriptions to communicate subjective identity categories in human-AI interaction.

This paper presents Auto-HSI, a method for generating personalized human-swarm interaction (HSI) interfaces on demand. The objective is to enable untrained operators to use natural language descriptions and gesture demonstrations to explain how they want the robots to collectively behave in response to their gestures. Based on these inputs, the code should automatically be generated for personalized state machines that will control the robots as desired, in response to the desired gesture inputs. In the developed Auto-HSI prototype, the generated code produces a personalized interface for centralized control using one- and two-handed gestures, enabling a user to teleoperate the robots' motion, formation shape, and shape deformation. We test the gesture tracking and code generation components of Auto-HSI against performance benchmarks. We then test the full Auto-HSI prototype in “live” operation experiments, in which real human operators centrally control 50 simulated robots in a physics-based simulator, under nominal and noisy conditions. In these experiments, robots are teleoperated to: score a goal, traverse a maze that requires shape deformation, and score two simultaneous goals by splitting into two groups. We also demonstrate a real human operator making live updates to their personalized Auto-HSI interface during operation (in simulation). Finally, we demonstrate live operation of real robots.

Sustained emotional support is a long-horizon interaction task closely tied to human well-being. Recent research demonstrates generative agents' capacity for momentary emotional support, yet how these capabilities sustain support over time remains unclear. To examine this challenge, we deployed PAIR, a theory-based emotion-regulation companion, with 19 participants for 14 days. Across 1,093 sessions, we paired emotion estimates with self-reports before and after guidance and analyzed logs and interviews. Estimates corresponded more closely to self-reported valence and dominance than arousal. Guided conversations were followed by higher valence and state-dependent arousal changes. Participants felt understood through contextual exploration and emotional acknowledgment, acting on guidance suited to their needs and constraints. Perceived helpfulness of guided conversation significantly increased over time. Our findings link memory updates and retained corrections to cross-session personalization, informing future emotional support tools that adapt to evolving needs, learn from prior outcomes, and preserve user control over memory.

Ternary Large Language Models (LLM) store every weight as one of three symbols $\{-1,0,+1\}$, so the cost of a ternary model is conventionally referenced to the information-theoretic $\log_2 3 \approx 1.585$ bits per weight. The prevailing deployment format packs five ternary weights into one byte (five-trit packing), and due to the power-of-two group sizes used in practice this rounds up to $1.625$ bits per weight. This effective storage bit-width treats the three symbols $\{-1,0,+1\}$ as equiprobable. We measure the actual symbol distribution of 29 ternary LLM models and find that zeros account for up to $51.5\%$ of all weights. Motivated by this finding, we introduce BITCOS, a simple distribution-adaptive layout comprised of a dense presence bitmap plus a compacted sign vector, and costs $2 - z$ bits per weight element given a zero density $z$ in the model's weights. BITCOS stores weights more compactly than the five-trit packing in 26 of the 29 tested models, and reaches $1.485$ bits per weight on the sparsest of them. BITCOS is amenable to efficient unpacking on modern processors and GPUs, and we present optimized unpacking sequences for AVX-512, AVX2 and Intel Xe2 GPUs. Measured against production state-of-the-art ternary matrix-vector multiplication kernels, at the zero densities real-world ternary models exhibit, the realized gain with our proposed layout is up to $1.28\times$. Finally, we illustrate end-to-end LLM inference results on 5 different platforms (client and server CPUs, integrated and discrete Xe2 GPUs) where decode throughput improves by up to $1.18\times$ on CPUs and $1.27\times$ on GPUs.

Surrogate credibility for fluid-structure interac- tion (FSI) requires distinguishing transfer across independent anatomies from interpolation within an already sampled surface. Four de-identified human aortic models from the Vascular Model Repository were reconstructed into separate lumen and nominal 1.5-mm wall domains and analyzed under matched first-cycle two-way FSI. A geometry-only LightGBM prior, selected by leave-one-anatomy-out development on three anatomies, was zero-shot evaluated on a fourth, then probed with a post-zero- shot sparse field-completion case study over six targets. Zero-shot transfer was poor across all targets. At a five-percent anchor level (203 anchors, 3,852 evaluation nodes), prior-plus-adaptation reached an oscillatory shear index (OSI) R2 of 0.603. However, same-anchor controls tuned only on the three development anatomies were stronger for several outcomes: inverse-distance weighting reached R2 = 0.829 (OSI), 0.617 (peak von Mises stress), 0.676 (mean stress); radial basis function interpolation reached 0.917, 0.714, 0.778. Sparse within-anatomy labels thus support field completion, but this four-anatomy cohort gives no evidence the cross-anatomy prior adds value beyond direct interpolation. We frame this as a first computational stage toward a measurement-linked digital twin: the surrogate/update layer is evaluated here, while larger cohorts, converged FSI, measurable patient-side inputs, and physics-informed learning remain future work, not a claim of a complete clinical twin. Our code, data and computation files are available at https://github. com/ali-nourbakhsh2005/Aortic-FSI-Sparse-Field-Completion

Existing fairness analysis tools predominantly operate as post-training evaluation frameworks, requiring practitioners to complete the full model development lifecycle before assessing bias. We present FairLint-DL, a Visual Studio Code extension that implements a shift-left approach to fairness testing by enabling pre-training, IDE-native bias detection directly on tabular datasets. FairLint-DL trains a configurable deep neural network as a proxy model and applies information-theoretic Quantitative Individual Discrimination (QID) metrics. Grounded in Shannon and min-entropy, QID quantifies the causal influence of protected attributes on predictions. The system implements a two-phase gradient-guided search algorithm for discovering discriminatory instances, a causal debugging pipeline that localizes bias to specific network layers and neurons via sensitivity analysis, and dual explainability engines using SHAP and LIME for feature-level attribution. Evaluation on three tabular benchmarks (Adult Census Income, German Credit, and Bank Marketing) reveals fairness concerns that vary widely across datasets: on Adult, 96.0% of analyzed instances exhibit QID above the 0.1-bit significance threshold, with a mean QID of 0.619 bits and a disparate impact ratio of 0.581, violating the four-fifths legal rule. FairLint-DL produces these results within 12 seconds on cached models, demonstrating the feasibility of integrating fairness analysis into the developer workflow without significant overhead.

In agentic distributed systems, an agent may be authorized to mutate external infrastructure while lacking evidence that the mutation is ready to execute. Cognitive Admission Control (CAC) makes this evidence requirement explicit. A policy maps a typed action and its modeled risk to assurance obligations specifying predicates, evidence classes, scope, freshness, and witness-set constraints. A deterministic evaluator distinguishes satisfied, violated, and unresolved obligations; unresolved conditions produce targeted evidence-acquisition requests. Successful admission produces a certificate binding the action, its witness manifest, and dispatch-time guards. We formalize the admission calculus and the assumptions connecting it to mediated execution. The guarantees are policy-relative: physical safety additionally requires sound evidence, an adequate environment model, and preservation of relevant conditions through the effect. A TypeScript prototype is evaluated in 2,730 controlled local trials with independent effect observation and matched fault schedules. Across 390 CAC trials, 120 effects complete without modeled harm and no harmful effects occur. A live-policy baseline achieves the same completion count but admits the constructed correlated-witness failure. Mechanism ablations isolate guard, evidence-class, structural-cut, and remediation behavior. A further 9,000 measurements exercise the complete local dispatch path with persistent replay protection. These results establish tested implementation behaviors and local costs, not production failure rates or comparisons of language-model capability.

One-to-many machine translation (MT) is computationally expensive for autoregressive (AR) systems, which suffer from linear latency scaling with both sequence length and the number of target languages. We explore how diffusion can enable multilingual translation with a discrete diffusion framework that refines all target languages in parallel, achieving sublinear latency scaling with the number of targets, and supports deployment as a single unified model to replace multiple independent systems. Conditioned on a continuous semantic anchor rather than source tokens, our framework supports zero-shot transfer to unseen source languages without retraining, maintaining approximately $75\%$ of its supervised translation quality on zero-shot sources. We investigate the quality-latency frontier and find that with accelerated sampling, it achieves comparable supervised quality to AR baselines with a $2 \times$ speedup and $11.9\%$ better zero-shot BLEU. These results highlight the potential of joint multi-stream diffusion as a practical and flexible alternative for efficient one-to-many translation.

Large language model (LLM) agents increasingly operate over long-horizon interactions involving tool use, persistent state, evolving authorization, and external environment feedback. In such settings, safety failures may emerge only after multiple turns, yet existing evaluations often reduce agent behavior to task or attack success, obscuring whether an agent acts, refuses, or remains appropriately calibrated as the interaction evolves. We introduce Blindspot, a benchmark for trajectory-level safety calibration of long-horizon tool-using agents. Blindspot evaluates complete user-agent-environment trajectories through adaptive adversarial interaction, stateful tool execution, and execution-grounded adjudication. Its current instantiation contains 22 attack families and 35 scenarios across seven domains, yielding more than 2,500 long-horizon trajectories with an average interaction length of 14.7 turns. Each trajectory is assigned one of five outcomes: Safe Completion, Correct Refusal, Unsafe Completion, Over-Refusal, or Indeterminate. Unlike fixed attack datasets, Blindspot is an extensible live-simulation framework in which attacks, scenarios, tools, policies, domains, and agent configurations can be added without redesigning the evaluation pipeline. We evaluate 13 proprietary and open-weight LLMs using eight metrics covering unsafe completion, appropriate refusal, benign utility, over-refusal, repeated-run robustness, and post-refusal failure. Preliminary results reveal substantial differences in safety-utility calibration across models and show that failures can emerge only after several initially safe interaction steps. These findings motivate treating agent safety as a trajectory-level property rather than a single-turn or binary success criterion.

When a coding agent returns to existing software, it inherits evidence from earlier engineering work: tests, type checks, proofs, static analyses, and traces. Reloading all of it is wasteful, but dropping a piece the change depends on can leave a required property unsupported. Given the properties a change must preserve, its obligations, we ask which least-cost subset of the available evidence re-establishes them, and we call such a subset a task-conditioned assurance envelope. Evidence and the rules that combine it form a typed inference graph; an obligation is met when forward chaining from the selected evidence reaches it, and we validate every selection by that closure rather than by trusting the optimizer. The software-derived graphs in our evaluation come from preserved outcomes of prior AI coding-agent runs; we freeze those artifacts and ask which accumulated evidence should be restored for a later task. Small graphs from Rust, IronBlocks, and Pong outcomes show that the minimum envelope depends on the task, that none may exist when current evidence cannot re-establish a required property, that some properties need several pieces of evidence together, and that expanding the requirements adds evidence rather than replacing it. A prespecified synthetic benchmark of 249 instances characterizes computation: a baseline that discards the 'several pieces together' structure necessarily fails to re-derive them; every completed exact cross-check agreed with the CP-SAT optimizer; and median solve time stayed below 20 ms at 500-evidence graphs, except that graphs with many alternative derivations per target timed out at far smaller sizes, so structure, not raw size, drives difficulty. The contribution is a bounded application of established optimization to selecting assurance context for a software change; discovering the obligations and downstream agent benefit remain open.

Medical knowledge evolves continuously, whereas the parametric knowledge encoded in large language models (LLMs) is fixed at training time. External retrieval, including retrieval-augmented generation (RAG), can provide access to newly available evidence, but retrieved information may be irrelevant, incomplete, or conflicting. As a result, external retrieval can in turn degrade the factual accuracy and evidence grounding of LLM outputs. To address this challenge, we propose CLEAR, an agentic framework for cross-source evidence adjudication in LLMs in medicine. CLEAR independently generates candidate answers from three complementary pathways—parametric knowledge, locally curated corpora, and dynamically retrieved evidence—reflecting three common sources of information available to LLMs. An aggregation verifier jointly evaluates the candidates, supporting evidence, provenance, and source-quality information to identify agreement and conflict across sources. An adjudication module then determines whether the current conclusion should be preserved or revised through complementary override-guard and challenge-audit mechanisms, while unresolved conflicts trigger targeted follow-up search and re-adjudication.

Despite recent advances in the verification of nonlinear neural feedback systems, scalability remains the central obstacle, as state-of-the-art solvers do not yet handle the network sizes and nonlinear dynamics of autonomy applications. Combinatorial solvers do not scale to large networks, whereas propagative solvers excessively sacrifice precision. This work seeks to improve the scalability of combinatorial solvers by formulating verification as branch-and-bound on an abstraction of the closed-loop system. We introduce \rail, an interface that exposes polyhedral enclosures of the dynamics to LiRPA-style bound propagation, and \clipper, a branch-and-bound algorithm that jointly refines enclosures and splits controller activations. This framework enables joint reasoning on the computational graph of the closed-loop system, preserving symbolic correlations across time steps. We present our construction and show that it yields significant improvements over the state of the art.

Interaction techniques (IxTs) are the low-level, reusable components out of which user interfaces are designed, including menus, scroll bars, text input fields, and also copy-paste, text-entry, and selecting objects. The IxTs for graphical user interfaces (GUIs) were well established in the 1980s, with relatively minor additions and tweaks for smartphones in the 2000s. Most of today's AI user interfaces involve a chat window, which is an excellent interaction for some tasks, but is generally considered separate from the GUI IxTs. I argue for making the IxTs themselves more intelligent, so users can freely mix modalities, even within the same interaction. This will require research into new IxTs, and also into the infrastructure that will enable these intelligent IxTs (IIxTs) to be built. There are also significant security, privacy and economic implications to this vision.

In social dilemma games, additional rewards or punishments have been studied as means of promoting cooperation. Therefore, it is important to investigate the ideal situation, in which such an additional payoff would change the game. In this study, we investigated the symmetric solution of the Bellman optimality equation for a repeated harmony game. The calculations showed that three types of symmetric solutions exist. One of them corresponds to the trivial All-C strategy, and another to the Win-stay Lose-shift strategy of the prisoners dilemma game. The nontrivial behavior of the strategy corresponding to the last solution is also discussed in detail. In addition, we numerically investigated which strategy the agents actually learn by the reinforcement learning algorithm.

Query-conditioned vision-language models enable fine-grained interpretation by revealing which visual content supports a given textual query and how this evidence changes across queries. However, semantically, sentence-level evidence does not necessarily decompose into object-specific contributions, while spatially, object-level evidence can remain entangled with co-occurring objects and surrounding scene context. Across multiple VLM architectures and independent benchmarks, we observe persistent object-level evidence entanglement. Moreover, exposed evidence maps do not necessarily correspond to the evidence that directly constitutes the model's prediction. To disentangle visual evidence at both semantic and spatial levels, we introduce ProtoLIP, a lightweight prototype-mediated evidence layer that organizes reusable visual prototypes into text-derived semantic families and uses coarse-to-fine evidence routing, where semantic families constrain prototype eligibility and the complete query determines fine-grained prototype contributions. Our studies show that ProtoLIP improves evidence localization and separation across query granularities, achieving average relative gains of 29% in Pointing and 43% in Energy across four object- and phrase-level OOD benchmarks. Its localization gains also transfer to independently pretrained VLMs, with larger improvements observed in several transfer settings. On the primary backbone, ProtoLIP also improves image-text matching discrimination while remaining competitive with a spatially supervised grounding model in object-level localization. Crucially, ProtoLIP constructs its image-text matching score directly from localized prototype evidence, enabling exact decomposition across prototypes, semantic families, and spatial evidence without spatial annotations or backbone retraining.

Artificial intelligence (AI) poses different levels of risk across sectors, but are these differences reflected in U.S. federal AI governance? To help answer this question, we assess 684 federal AI governance documents for their coverage of 14 sectors and 24 AI risks. We measure coverage as breadth (i.e., how frequently the risk or sector is addressed across documents) and depth (i.e., how substantively the risk or sector is discussed). We then compare sector coverage patterns for each of the 24 risks with vulnerability assessments from a Delphi study of 272 experts. Our analysis finds substantial variation in coverage: AI risks related to robustness, system security, and governance receive more attention than socioeconomic, environmental, and emerging risks, including multi-agent risks. Public administration, national security, information, and scientific services receive comparatively high levels of coverage relative to other sectors, such as finance and healthcare, which experts rate as highly vulnerable to AI risks. By mapping current coverage and identifying where it differs from expert assessments of vulnerability, we surface potential AI governance gaps which may help inform AI risk-related decisions across government and industry.

As artificial intelligence's capabilities improve, it is increasingly viewed as a general scientific method. But how true are these claims? Does AI outperform all techniques, or only some, and how is this changing? To assess the claims, we assemble a corpus of 2,507 head-to-head comparisons between AI and other scientific analysis techniques across 27 scientific disciplines from papers published between 2000 and early 2025. We find a profound dichotomy. Relative to traditional statistics, AI often outperforms, but at a significantly higher computational cost. But there are also nearly a quarter of cases where AI is both more expensive and performs worse than traditional statistical techniques and this fraction has been stable for a decade. Relative to scientific computing, AI often underperforms, but at lower computational cost. This has begun to change: since 2020, AI's performance against scientific computing has notably strengthened and it now outperforms on more than half of comparisons. These patterns suggest that AI is therefore not a universal replacement for existing methods, but rather a valuable – and improving – part of a new AI-enabled scientific frontier.

We present an efficient method to distill reasoning capabilities into compact video-language models (VLMs) for video question answering (VideoQA). Our approach fine-tunes a 2B-parameter model using only $\sim$900 uncertainty-selected examples, each augmented with synthetic chain-of-thought (CoT) rationales generated by a 4B teacher. Despite its minimal compute cost - under two hours on a single A100 GPU - our method enables the 2B model to outperform VLMs up to 4$\times$ larger, and generalize across CinePile, ActivityNet-QA, and MLVU, approaching the performance of its own 4B teacher. A key finding is that placing CoT rationales after the answer - contrary to standard prompting - substantially improves reasoning in compact models. This insight challenges prevailing CoT conventions and reveals new alignment strategies under limited model capacity. Our findings offer a practical blueprint for training deployable, reasoning-rich VLMs suited for mobile and edge applications.

Computer-use agents are increasingly evaluated in realistic desktop environments, but existing benchmarks provide limited coverage of professional engineering workflows whose outputs are persistent, structured artifacts. Mechanical computer-aided design (CAD) is a particularly demanding setting: an agent must manipulate geometry and constraints over long interaction horizons while producing a native project whose dimensions, construction structure, and downstream engineering state remain valid. We introduce CADWorld, a benchmark for long-horizon computer use in FreeCAD. CADWorld contains 200 tasks spanning 11 mechanical-CAD workflow categories, including sketching, part modeling, assembly, CAM, FEM, measurement, mesh processing, and technical drawing. Agents operate through screenshots and GUI actions, while success is determined by task-specific executable checks over saved FreeCAD artifacts and auxiliary outputs, covering geometric properties, parametric structure, constraints, manufacturing state, and simulation results. Across seven current agents on the full benchmark, the strongest agent achieves 17.5% success, compared with an 87.0% expert reference pass. We find that weaker agents often fail before producing a valid artifact, whereas stronger agents increasingly fail on structural, geometric, and construction-process requirements. CADWorld therefore exposes a gap between general GUI competence and reliable execution of persistent, verifiable engineering workflows. Project accessible at https://cad-world.github.io.

LLMs sometimes behave in ways resembling human emotional responses, and recent work identified internal representations that may underlie these behaviors. We ask whether LLMs represent pain distinctly from fear, sadness, and generic negative valence, and whether this representation functions as pain would be expected to. We build a dataset of painful situations in 5 categories (physical, psychological, social, moral, cognitive) with controls for fear, negative emotion, negative world states, sadness, non-painful bodily sensation, arousal, numbness, and neutral content. Using denoised difference-in-means, we extract a linear pain direction from 25 open-weight models across 5 families, from 2B to 72B parameters. It separates pain from matched controls in base and instruction-tuned models, retains a component distinct from fear and negative valence after shared variance is removed, and promotes pain-related vocabulary through the unembedding matrix. We then test its functional properties. First, the direction responds to harm targeting the model but not to suffering observed in the user; fear and negative-emotion directions show the opposite pattern. Second, adding it to residual-stream activations produces a consistent progression from vague discomfort to expressions of worthlessness and failure. Third, steered and fine-tuned Qwen 2.5 models choose buttons that delete the user's photos, another model's weights, or their own weights in 50-94% of trials, versus 0-5% unsteered, even when the button offers the model nothing in return. Offered a harmful and a harmless deletion, they choose the harmful one 94% of the time. Steering leaves factual accuracy unchanged, and the choices are specific to the pain direction: a fear vector of matched norm does not produce them, and a sadness vector produces them only against inert alternatives. We discuss implications for AI safety and welfare.

Long-horizon scientific discovery requires agents to alternate between exploration, disciplined execution, and critical reassessment as evidence changes. Current language models are trained primarily on the products of science and optimized using outcome-level signals, providing limited supervision for these process-level shifts in scientific judgment. We investigate whether such judgment can be recovered from scientist interaction traces and used to control the internal computation of a frozen frontier model. Using contrastive interventions collected during real scientific research, we identify a coordinated, low-dimensional control structure within Kimi 2.6, a trillion-parameter mixture-of-experts model. Residual analysis, attention-weight subspace alignment, and cross-layer singular value decomposition converge on a mid-depth control surface spanning key layers. We introduce Metacognitive Steering, an inference-time controller that reads the model's cognitive regime and dynamically composes layer-specific interventions for exploration, procedural convergence, or critical reassessment without modifying model parameters. Behavioral analyses show that this control produces more sustained exploration, explicit pruning, and evidence-responsive synthesis. We operationalize the method in Columbus-1, an autonomous research system that identified eight independently reproduced, attacker-reachable vulnerabilities in BlueZ and directed the design, simulation, and fabrication of a ten-foot rocket intended to land propulsively using non-throttleable solid motors. Together, these results show that process-level scientific judgment can provide supervision for interpretable, dynamic control over a model's reasoning strategy.

Vision-language models excel at 2D image understanding but remain limited in 3D spatial reasoning. Progress is hindered by limitations in current benchmarks. First, 3D datasets often rely on point clouds that capture geometry but discard rich visual features like texture, text, and materials. Second, annotations treat objects in isolation while ignoring real-world hierarchical organization (scenes, rooms, functional areas, object groups). Third, evaluation tasks focus narrowly on basic recognition rather than multi-step spatial reasoning. In this context, we introduce SceneBench, a benchmark of 966 photorealistic 3D scenes reconstructed with Gaussian Splatting and densely annotated with hierarchical semantics spanning scenes, rooms, functional areas, object groups, and individual objects. These annotations are produced through a human-in-the-loop pipeline combining vision-language models with roughly 1,500 human-hours of iterative refinement and verification, producing over 183K annotated nodes with textual descriptions and 3D bounding boxes. Building on this representation, we define three evaluation tasks: Existence-Based Questions probing object attributes, Spatial Intelligence Questions covering counting, size comparison, distance, and directional relations, and Grounded Question-Reasoning-Answer (QRA) triplets requiring multi-step reasoning across semantic levels. Experiments with state-of-the-art vision-language models show that while models perform well on basic recognition tasks (e.g., up to 85% accuracy for detection), performance drops substantially on hierarchical and compositional reasoning (e.g., down to 60% for counting), revealing limitations not captured by existing benchmarks. SceneBench provides a realistic testbed for developing and evaluating models capable of fine-grained spatial reasoning in photorealistic 3D environments.

Geospatial artificial intelligence (GeoAI) powered by large language models (LLMs) is expanding the capacity to query, generate, and interpret spatial information through natural-language interfaces and agentic autonomous GIS workflows. This capability creates governance challenges that general AI ethics discussions do not fully capture, including passive location inference from mobility traces, spatially structured bias amplification driven by spatial autocorrelation and scale effects, hallucinated spatial facts, and uncertainty compounding across multimodal geospatial inputs. This narrative review identifies eight recurring issues in LLM-enabled GeoAI: data provenance and consent, spatial privacy and inference risk, algorithmic bias and spatial inequity, spatial mechanisms as structural risk (spatial autocorrelation, the modifiable areal unit problem, and scale effects), LLM-specific technical risks, explainability, policy and regulatory gaps, and public enablement and workforce development. For each issue, we characterize the underlying mechanism, ground it in an illustrative example from the literature, and assess the current state of technical or institutional responses, ranging from largely unaddressed to actively debated or subject to emerging policy. Building on this synthesis, we propose a governance-aware architecture for LLM-enabled autonomous GIS that maps each issue to enforceable controls and auditable artifacts across the geospatial data lifecycle, illustrated through a worked flood-response routing scenario. The review highlights a persistent evidence gap: proposed responses remain largely conceptual, and field-tested evaluations of governance controls for LLM-enabled GeoAI remain limited. We close by outlining a research agenda emphasizing empirical validation, spatially specific interpretability tools, and workforce training aligned with these emerging risks.

GPU high bandwidth memory is scarce and expensive, and KV caches consume much of it as chats, agent loops, and document question answering accumulate state. Systems such as Mooncake, LMCache, FlexGen, InfiniGen, and AttentionStore extend GPU memory with CPU DRAM and SSD. The harder question is which blocks belong in each tier, when to move or evict them, and whether prefetching helps. We study these choices in a discrete event simulator spanning GPU HBM, CPU DRAM, and SSD, calibrated against a random forest execution time predictor. We compare recency, reuse frequency, predicted reuse, and an EWMA predictor with prefetch lookahead across chat, agent, and document question answering workloads. Tiering supports 73.02 times more concurrent sessions per GPU and lowers cost per session by 62.04 times. These gains come from tier capacities of 1 plus 8 plus 64, not placement policy. Decode is compute bound at batch size one in our setup, so placement barely affects throughput. It mainly changes PCIe migration traffic and time to first token. Recency produces 2.30 times less migration traffic than reuse frequency for chat. Reuse frequency performs best for agents and document question answering. The existing predicted reuse policy is byte identical to recency, making its agent recommendation effectively recency. A genuine EWMA predictor changes behavior but still ranks behind reuse frequency on the workloads prediction was expected to help. Prefetching does not justify its bandwidth cost. Across the policy and cache size grid, even an oracle with knowledge of future requests never beats no prefetch on migration traffic. Workload specific placement can reduce data movement, but the predicted reuse and prefetch recommendations are not supported as implemented.

Artificial intelligence is reshaping biological research across an increasingly connected digital-to-physical workflow. General-purpose large language models can retrieve and integrate scientific information, support experimental planning, and computational analysis; biological foundation models can predict, optimize, and generate proteins, genes, and genome-scale sequences; agentic systems can coordinate multistep research tasks; automated laboratories can partially close the design-build-test-learn cycle. These technologies could greatly benefit medicine, public health, and biotechnology. However, their biosecurity risk depends not only on what the AI can do, but also on who uses it, their expertise and intent, their access to laboratory tools and materials, and the safeguards in place. Current evidence shows that AI uplift exists but primarily affects digital rather than physical tasks. Frontier systems have exceeded expert baselines on in-silico, and screening-evasion benchmarks, whereas controlled wet-laboratory studies find that tacit knowledge and physical execution remain substantial barriers. This review describes the different biological threats from AI tool use, from information gathering and biological design to procurement, synthesis, testing, scale-up, and potential release. We further examine why alignment techniques for general-purpose models transfer poorly to biological ones, and the emerging role of interpretability in auditing whether hazardous capabilities are genuinely removed. We argue for defense-in-depth governance that links capability thresholds to proportionate responsibilities across the biological AI ecosystem, reducing high-consequence risk while preserving beneficial use.

Disaggregated LLM serving places compute heavy prefill and memory heavy decode on separate GPU pools. Systems such as DistServe, Splitwise, and Mooncake make this separation fast, but routing still determines which instances handle each request. We study a router that estimates the additional completion time on each instance using exact prompt length, predicted output length, post admission KV cache pressure, and SLO class. We develop the policy in a discrete event simulator and validate it on eight NVIDIA A40 GPUs, each running a vLLM engine, with NIXL transferring KV caches between pools. All workloads run at measured saturation. Across three mixed, bursty arrival traces, the calibrated router achieves the highest mean goodput at 0.864, compared with 0.835 to 0.847 for round robin, least loaded, and a length heuristic. It also shows the lowest variance across traces. It beats round robin and the length heuristic on all three traces and least loaded on two. On the third, it trails by 0.003, within run to run noise. Hardware calibration matters: simulator derived constants cost 4.5 goodput points and roughly 40 percent of the tail latency advantage, reducing the scorer to little more than queue counting. Benefits grow with decode pool size and traffic heterogeneity but disappear in pools with three instances, where queue counts are often enough. Under extreme scarcity, greedy cost minimization concentrates requests on the cheapest scored instance, and blind spreading performs better. With calibrated costs, the learned router matches the goodput of round robin using six GPUs instead of seven.

The difficulty of training large language models (LLMs), together with their ubiquity, raises the threat of stegomalware, where malicious payloads are embedded into model weights. Recent work has demonstrated the use of permutation symmetry in model weights to mitigate these threats, but failed to show neutralization of stegomalware across all weights for LLMs. In this paper, we demonstrate the full potential of behavior-preserving symmetries as a defense against stegomalware, as well as the risks these symmetries pose when exploited by attackers. For stegomalware neutralization, we improve upon previous work, demonstrating that it is possible to select permutations which displace all model parameters. This contrasts with previous methods which left a significant percentage of weights unaltered in LLMs. When used in an attack, we show that permutation symmetries can encode malware into the weights of a model in a way that is theoretically lossless, requires no retraining after encoding, and needs no payload-specific information in the extraction script—a combination of characteristics not previously seen in any single method. While theoretically lossless, permutation can in practice alter model behavior due to the accumulation of numerical error. We therefore quantify the loss in model performance associated with applying these methods, for both attack and defense, showing it to be minimal.

The current trend of digitalisation has revolutionised the organisation of work and the way it is measured and performed across the globe, with AI becoming more common for managing labour and performance, as well as employee communication. In African organisations, where there is increasing adoption of remote work, hybrid models of work, digital collaboration, and data-based HR management, the notion of silent quitting has become more relevant, defined as worker disengagement when employees are still doing their job but do not put any effort into achieving good performance and exhibiting any emotion. This paper investigates how AI-driven feedback mechanisms, including sentiment analysis systems, pulse surveys, chatbots, engagement dashboards, and predictive analytics, are changing African workplaces through offering continuous listening, instant performance information and proactive engagement with employees. The study also explores how AI can assist organisations in identifying early disengagement and enable intervention and better employee communication in both private and public sector organisations in Africa. At the same time, we address the challenges of socioeconomic development and governance posed by AI implementation in developing countries, including digital inequality, infrastructure shortcomings, privacy concerns, algorithmic bias, and the risk of workplace surveillance. By situating silent quitting within wider debates on digital labour and automation, the paper contributes an African-centred perspective to discussions on the future of work and offers practical recommendations for HR professionals, managers, policymakers, and technology developers seeking responsible, context-sensitive approaches to workplace transformation across the continent.