Papers

Solving complex tasks in domains such as science, medicine, law, and finance often requires assembling interdependent information scattered across vast, heterogeneous sources far beyond model context limits. Existing approaches tackle this challenge by organizing information into more manageable representations over which models can reason, such as graphs, textual memories, and retrieval collections. These representations dictate what downstream reasoning is possible and, ultimately, whether it succeeds; yet their design and construction remain largely ad hoc. In this work, drawing on the cognitive theory of relevance realization, we propose concrete principles for designing AI systems that construct effective representations of very large contexts. We analyze existing approaches and show how their successes and failures can be explained by their alignment with these principles, and introduce R3Con, a harness that systematically operationalizes them. We evaluate R3Con against nine state-of-the-art baselines on two recent benchmarks of reasoning over large document corpora. On these benchmarks, R3Con substantially outperforms the strongest baseline, by $20$ and $8.4$ percentage points. It also enables smaller models to outperform much larger ones: R3Con with 4B and 9B models outperforms all evaluated 35B baselines, while R3Con with Qwen3.5-35B-A3B outperforms Claude Code with Claude-Sonnet-5 at $3.7\times$ lower cost. Our results show that context representations following our principled approach can reduce reliance on model scale, pointing toward a future of AI systems with frontier-level performance powered by smaller models.

We study exact pattern matching on indeterminate strings, where a text or pattern position may represent a set of symbols rather than a single letter. Focusing on Boyer-Moore-style methods, we present new bad-character rules (BC Rules I-IV) and a new good-suffix procedure, computed by Fast_GSR_Indet_Shift, which avoids the per alignment recomputation used in BM_Indet [12] by shifting with a single preprocessed position-indexed table. We conduct a systematic experimental evaluation of sixteen algorithms, including classical bad-character adaptations (e.g., Horspool, Sunday, and Zhu-Takaoka) and hybrids that combine these bad-character rules with Fast_GSR_Indet_Shift. Across synthetic scaling experiments and a case study on the E. coli K-12 MG1655 genome, the Fast_BM_Indet hybrids consistently outperform BM_Indet and KMP_Indet [12], in some settings by up to two orders of magnitude. We also find that Zhu-Takaoka is the strongest bad-character-only adaptation on small alphabets and genomic data, while the Fast_BM_Indet variant using BC Rule I offers comparable performance, making it attractive for larger alphabets. We conclude with practical guidance on choosing among these variants for indeterminate string applications.

A graph is 1-planar if it has a drawing in which every edge is crossed at most once. We show that the smallest cubic non-1-planar graphs have $30$ vertices. Two such graphs are the Tutte-Coxeter graph of girth eight and a graph of girth seven that we call the Byte graph. Every subcubic graph with fewer than $30$ vertices is 1-planar. Our proof is computer-assisted, but directly testing all relevant graphs is impractical. To establish non-1-planarity of the two graphs, we extend a SAT-based solver with a custom clause propagator based on separating cycles and a case split based on graph automorphisms, allowing independent cases to be solved in parallel. To show that all smaller subcubic graphs are 1-planar, we introduce the concept of $k$-flexibility: every set of at most $k$ prescribed edges can remain uncrossed in some 1-planar drawing. We use this property to reconstruct 1-planar drawings of larger graphs from drawings of smaller $k$-flexible graphs. This replaces exhaustive testing of more than forty billion cubic graphs with computations on far fewer graphs of smaller order.

Background. A connectome specifies the substrate available to a nervous system, not how much of it a given computation requires. Reported compact sufficient circuits are point estimates: one subset, one task, one fidelity threshold, one parameterisation. Methods. We formalise computational sufficiency as a cost-indexed family of neuron subsets conditioned on a task and a tolerance pair (permitted degradation epsilon, permitted failure fraction rho), separate task adequacy from phenotype fidelity, and certify candidates on held-out parameter draws with one-sided bounds rather than reporting a single minimum. We measure three quantities a single circuit cannot express: non-decomposability, degeneracy, and an intervention-derived role taxonomy. Results. On a released walking simulator, the published 148-neuron circuit is certified sufficient with margin (mean Dtask = 0.017 over 32 held-out draws, no failure), no 146- or 147-neuron subset of it is sufficient, and removing any one of its three interneurons destroys adequacy. Complement splitting recovers the published set exactly in three of seven seeds and reaches its size in five, within 66-97 evaluations, against 221 neurons for the bundle hierarchy at the same budget. Varying only the task changes which neurons are needed: our best-found solutions retain 155 and 149 neurons and share only 3 of 11 searchable neurons. Relaxing the tolerance reveals alternative sufficient circuits, and the whole-CNS interface alone fails every replicate while 93 search-chosen neurons give a certified solution. Conclusions. Sufficiency is set-valued and conditioned jointly on task and tolerance, and no single search strategy dominates across regimes. A frontier is interpretable only alongside its interface-only baseline, its tolerance and its budget. Every result here is model-relative.

Probabilistic forecasts of day-ahead electricity prices are increasingly used to manage price risk, yet competing models are usually ranked on aggregate scores that average over the price regimes in which uncertainty matters most. This study evaluates probabilistic forecasts of hourly EPEX SPOT GB day-ahead prices over a full held-out calendar year, applying a common pre-auction information boundary and rolling-origin design to linear quantile regression, a quantile regression forest (QRF), a LEAR-QRA benchmark and an encoder-decoder quantile long short-term memory (LSTM) network. Two routes to more reliable intervals are then compared: conformal recalibration and Quantile Regression Averaging (QRA). Aggregate scores separate the LSTM and QRF only weakly, with a non-significant 4.4% difference in continuous ranked probability score (CRPS), but their conditional behaviour differs sharply. Above 200 GBP/MWh the LSTM covered 0.12 of outcomes and QRF 0.71, consistent with the forest retaining realised spikes in its leaf distributions while the network learned a smooth upper tail. Conformal recalibration corrected marginal coverage but did not improve the interval score or materially repair spike coverage. QRA combining the LSTM, QRF and LEAR-QRA forecasts reduced CRPS to 5.14 GBP/MWh and raised coverage above 200 GBP/MWh to 0.65 at unchanged interval width. Recalibration and combination therefore address different failures and are not substitutes.

Bitcoin perpetual futures trade continuously with shallow liquidity and high price impact, making them prone to millisecond-scale flash crashes. Agent-based limit order book simulators are a natural tool for studying such events, but their behavioral parameters are usually hand-crafted or require full order book histories that are rarely available at scale. We propose a four-stage pipeline that constructs a reproducible 100 ms BTCUSDT flash-crash event panel from 66 detected events, elicits behavioral priors for four agent archetypes using an offline large language model conditioned on empirical microstructure anchors, instantiates a heterogeneous limit order book simulator with the validated priors, and applies causal discovery plus counterfactual analysis to the resulting synthetic panel. Across 500 simulation runs, LLM-elicited priors produce a crash rate of 0.176 versus 0.008 under uninformative uniform priors and generate heav- ier return tails. In the frozen counterfactual analysis, reducing leverage amplification by one within-sample standard deviation lowers the predicted crash rate from 0.176 to 0.136, whereas reducing directional pressure yields only a marginal change. These find- ings suggest that empirically anchored LLM elicitation can improve simulator fidelity and help identify plausible crash-amplification mechanisms in a controlled synthetic environment

Legged robot control is not yet practical in non-inertial environments, where the ground and surroundings accelerate in the inertial Earth frame, mainly because of the unknown, time-varying, persistent perturbations caused by environmental motion, compounded by the associated hybrid robotic dynamics. We formulate the problem of robot walking control as unknown-disturbance rejection for hybrid linear systems subject to hybrid sinusoidal disturbances (i.e., a linear combination of sinusoidal components with instantaneous phase jumps at discrete events). First, necessary and sufficient conditions for solvability of the disturbance rejection problem are derived and used to characterize the control input that rejects the disturbances. Second, with the Youla-Kucera parametrization, an output feedback controller capable of rejecting the disturbance is introduced. Third, an adaptive law is proposed to estimate the controller parameters that reject the disturbance in the practical case where the disturbance and its parameters are unknown. Experiments on bipedal walking during ground motion demonstrate substantially improved body-position tracking accuracy, with an average improvement of over 60% in simulations and up to 47% in real-world experiments.

Reliable out-of-distribution generalization remains a central challenge in machine learning, particularly when models are deployed in data-scarce high-fidelity domains. Existing approaches often rely on statistical distribution alignment, or invariance assumptions that can fail when domain shifts reflect changes in underlying mechanisms. Inspired by a longstanding paradigm in physics that understands complex systems by starting from tractable idealizations and progressively incorporating additional mechanisms and levels of fidelity, we propose physics-transfer (PT) learning, a physics-constrained learning theory for zero-shot generalization along a ladder of complexity. PT constructs chains of anchor domains (CADs) that connect low-cost source domains to complex targets and introduces computable feasibility and reliability indicators for assessing transfer before target-domain labels are observed. Under physical-consistency and stable-representation assumptions, we derive an operational upper bound on target risk in terms of source empirical risk, the PT indicators, and finite-sample errors, and establish convergence as adjacent complexity gaps shrink and data coverage improves. We evaluate PT on excited-state energy prediction in quantum systems and dynamical-transition prediction in classical systems using neural networks trained without target-domain labels. Across substantial distribution shifts, PT yields accurate high-fidelity predictions, while the indicators identify incorrect physical mappings before target-domain prediction fails. These results provide a model-agnostic framework for designing, diagnosing, and controlling generalization in data-scarce learning problems.

Distributed mixture-of-experts (MoE) inference is a promising architecture for deploying large language models (LLMs) at wireless edge networks because sparse experts can be placed across coordinated base stations (BSs), while the anchor node and user equipment (UE) can offload LLM inference tasks to BSs. The communication bottleneck is the MoE aggregation, where the anchor BS must recover a weighted sum of selected expert outputs before each decoding step. Over-the-air computation (AirComp) is well matched to this operation because the wireless multiple-access channel naturally superposes simultaneous transmissions. However, conventional AirComp minimizes communication distortion, whereas MoE aggregation errors have unequal impact on LLM outputs. We propose a logit-aware MIMO AirComp framework that estimates local logit sensitivity as block-level weights and jointly optimizes receive combiners and BS precoders under per-BS power constraints. We also develop an alternating algorithm that protects decoding decisions from aggregation perturbations. Using OpenCompass, we evaluate Qwen3-30B-A3B-Instruct-2507-FP8 on GSM8K and ARC-Challenge. At 30 dB aggregation SNR, perturbed Qwen3 retains 99.3% of clean GSM8K accuracy and 94.8% of clean ARC-Challenge accuracy. In a direct ARC-Challenge closed-loop audit over 295 examples, SW-AirComp achieves 43.73% accuracy at 30 dB and 26.78% at 20 dB, outperforming unweighted, matched-filter, and zero-forcing AirComp. In the wireless simulator, SW-AirComp reduces decision-relevant aggregation distortion; at 20 dB, its final weighted-sum mean-squared error is 53.5% lower than unweighted AirComp under the same channels, samples, and sensitivity weights. Logit-RMSE and logit-gap diagnostics further show that the gain comes from reducing output-logit perturbation and lowering the risk of top-token changes.

Let $q$ be a power of an odd prime $p$. We prove the MDS conjecture over $\mathbb F_q$ in every dimension $k$ satisfying \[ 2\leqslant k\leqslant B(p,q) \quad\text{or}\quad q+2-B(p,q)\leqslant k\leqslant q, \qquad B(p,q)=\left\lfloor\frac{(p-2)q+6p-10}{2p-3}\right\rfloor. \] For fixed $p$, the first interval is linear in $q$; together with duality, the theorem covers an asymptotic proportion $1-1/(2p-3)$ of all dimensions. The proof rests on a vanishing theorem for determinant relations over an arbitrary field of characteristic $p>0$. It yields full row rank for Chowdhury's matrices over a larger range of arc sizes. In particular, at $|G|=2k-3+n$ it proves Chowdhury's full-row-rank conjecture without the $q$-dependent restriction. Specialization to $\mathbb F_q$, together with the Ball–Lavrauw construction, gives the stated MDS range. We also prove that, for every odd prime power $q$, every normal rational curve in $\mathrm{PG}(N,q)$ is complete for $2\leqslant N\leqslant q-2$, and every projective Reed–Solomon code of length $q+1$ and dimension $2\leqslant k\leqslant q-2$ has covering radius $q-k$.

The SWOT (Surface Water Ocean Topography) mission is currently providing unpreceded high-resolution measurements of Sea Surface Height (SSH), revealing ocean features at finer scales. Nevertheless, the two-dimensional observations of KaRIn altimeter of SWOT suffer from instrumental errors. This noise degradation is altering the high frequencies of SWOT signal, and the small to sub-mesoscale dynamics of interest for some oceanographers. For this reason, Tréboutte et al. (2023) have developed a convolutional neural network (CNN) based on U-Net architecture to separate the noise from the physical signals contained in the SSH. Their approach has demonstrated great potential on simulated SWOT measurements, and Dibarboure et al. (2024) report a positive influence on actual flight data from SWOT. However, degraded denoising performance, and occasional negative side-effects have been observed in atypical conditions (e.g. very high surface waves, internal tides solitons). In this study, we illustrate some of these limitations and we present an improved approach of the CNN-based denoising: we modified the training procedure to obtain a more robust version of the algorithm, to avoid biases and artifacts in the denoised SSH. This paper also presents a more complete validation process with a robust and standardized evaluation benchmark: these metrics could be of interest to assess other SWOT filtering and denoising algorithms.

We prove a capacity criterion for fixed-degree spatially coupled MacKay–Neal codes on additive symmetric Markov noise. The noise bit flips with probability $p$, so the channel has one parameter and capacity $1-h_2(p)$. For every integer $\ell\geq4$ with $3/\ell<1-h_2(p)$, matched finite-window BCJR and sum-product decoding admit code sequences, possibly depending on $p$, with actual rate tending to $3/\ell$ and vanishing average information-bit error. The proof parallels the GEC-to-BEC argument: a conditional-entropy correction transfers BMS fixed-point positivity to the channel with memory. The correction also covers a capacity deficit of the effective BMS channel. A two-state posterior contraction supplies the additional derivative bound needed for threshold saturation. The result uses the cited BMS positivity theorem, with an analytic proof for $\ell\geq33$ and exact interval certificates for $4\leq\ell\leq32$.

Building automation systems are increasingly represented as semantic knowledge graphs (KGs) using ontologies such as Brick and ASHRAE 223P, creating a machine-readable substrate for artificial-intelligence applications. One promising application is translating natural-language questions into SPARQL (text-to-SPARQL), which would let building operators query these graphs through language agents, but progress is limited by the scarcity of large natural-language/SPARQL benchmarks. This paper presents Build2SPARQL, a large-scale benchmark for building KGs generated by a KG-grounded pipeline: SPARQL queries are produced and validated entirely by graph-traversal code, while large language models generate only the natural-language questions, keeping query correctness independent of model behavior. The pipeline mines six query-pattern families – linear chains, branching, UNION, aggregation, OPTIONAL, and attribute-filtered – and phrases each query across five vocabulary registers. Applied to 201 building KGs (180 Brick, 21 ASHRAE 223P), it yields 6,136 executable SPARQL queries and 30,680 questions. A two-rater human validation of 300 questions found 98.8% semantic fidelity, 98.8% naturalness, and 84.0% operational plausibility. A retrieval-augmented evaluation across three open-weight language models raised exact-match accuracy from 0.2-20% (zero-shot) to 56-65% (three-shot retrieved).

As ever larger language models grow more ubiquitous, it is crucial to consider their environmental impact. Characterised by extreme size and resource use, recent generations of models have been criticised for their voracious appetite for compute, and thus significant carbon footprint. Although reporting of carbon impact has grown more common in machine learning papers, this reporting is usually limited to compute resources used strictly for training. In this work, we propose a holistic assessment of the footprint of an extreme-scale language model, Noor. Noor is an ongoing project aiming to develop the largest multi-task Arabic language models – with up to 13B parameters – leveraging zero-shot generalisation to enable a wide range of downstream tasks via natural language instructions. We assess the total carbon bill of the entire project: starting with data collection and storage costs, including research and development budgets, pretraining costs, future serving estimates, and other exogenous costs necessary for this international cooperation. Notably, we find that inference costs and exogenous factors can have a significant impact on total budget. Finally, we discuss pathways to reduce the carbon footprint of extreme-scale models.

In recent years, disinformation has increasingly proliferated across social networks. The firing of fact-checkers from Meta and the disbanding of Twitter's Trust and Safety Council suggest that this trend will continue to escalate. While disinformation sources (e.g., social network accounts) can sometimes be identified, new accounts emerge daily, making tracking a moving target. In this work, we propose a graph-based methodology to discover previously unknown Telegram accounts that spread disinformation. Starting from a set of verified disinformation groups, we examine their interconnections and identify new accounts that contribute to the dissemination of false narratives. Our approach is language-agnostic, as it relies solely on structural relationships between accounts rather than analyzing their message content. Using this approach, we identify 37 previously unknown disinformation channels (a threefold increase). We demonstrate that Telegram channels display extremely dogmatic behavior with up to 80% of messages being labeled as propaganda. Our findings reveal that misinformation on Telegram spreads within tightly interconnected clusters, in some cases, with over 86K identical messages being shared to multiple channels, suggesting coordinated disinformation campaigns.

Is the same data equally useful to every learner? The question is whether the value of a training sample belongs to the sample or arises in its interaction with the learner. We test it with a selection experiment. Under a per-class budget, two kinds of samples compete for the slots: easy samples, which a short training run learns first, and coverage samples, which spread over the class in feature space. Easy samples win at small budgets and coverage samples at large ones, and the budget at which the winner flips, the crossover m0, can be measured. We ask whether m0 is set by the data or by the learner. To separate the two, we freeze the selected subsets and change only the learner that trains on them. On low-resolution ImageNet-100, doubling the width of a ResNet-18 moves the crossover from 57 to 85 samples per class on the same subsets. Enlarging the input grid from 32 to 64 pixels at fixed image information moves it from 85 to 57, and a single stride change reproduces or removes this shift. The grid shift shrinks as width grows. Under the native ImageNet-1k protocol, a 4x range of width moves the boundary by at most a few samples per class. On a Vision Transformer, coverage wins at every measured budget, even with easy subsets that work on the convolutional learner. The budget at which one kind of sample stops being the more useful kind therefore belongs to the learner, and the preferred regime of a selection strategy has to be read against the learner that will use the data.

Real-time motion generation for tendon-driven continuum robots requires accurate modeling of nonuniform bending and whole-body collision avoidance. This paper presents a unified actuation-space framework for planar multi-segment tendon-driven continuum robots. An energy-based variable-curvature model captures spatially varying tendon spacing and bending stiffness and provides analytical Jacobians for differential inverse kinematics and safety monitoring. A multipoint CBF-QP enforces backbone clearance under obstacle motion and actuation-velocity bounds, while its decision dimension depends only on the number of independently actuated segments. The model closely agrees with GVS references, with a maximum curvature error of $5.223 \times 10^{-2}\,\mathrm{m}^{-1}$. Over 100 MuJoCo trials, the proposed method achieves collision-free success rates of 96% and 100% in static and dynamic scenarios, respectively, compared with approximately 60% and 80% without CBF constraints. Hole-traversal tests further demonstrate safe motion in constrained environments. With 600 backbone monitoring points, the mean control-step time is 6.66 ms, demonstrating real-time whole-body safe motion generation.

This paper develops a cost-informed aggregation framework that learns an aggregate flexibility set of building heating, ventilation, and air-conditioning (HVAC) loads to minimize the aggregator's dispatch cost. Existing aggregation methods mainly use volume-oriented objectives and treat flexibility aggregation and downstream utilization as separate stages. Consequently, the resulting aggregate set may fail to preserve the flexibility most valuable for reducing downstream dispatch costs. To address this limitation, we represent aggregate HVAC flexibility using a parameterized storage-form surrogate and jointly learn the surrogate parameters and the inner-approximation objective from downstream dispatch-cost feedback. This cost-informed feedback allocates the surrogate's limited representation capacity to cost-relevant regions of the aggregate flexibility set. To account for electricity-price uncertainty, we formulate a distributionally robust conditional value-at-risk (DR-CVaR) downstream problem that captures both distributional ambiguity and tail risk. For efficient learning, we reformulate the DR-CVaR problem as a convex second-order cone program. We estimate cost gradients using randomized smoothing and a score-function estimator, avoiding differentiation through large-scale building-level optimization problems. Case studies using NYISO price data show that the proposed framework reduces dispatch costs relative to a volume-oriented aggregation benchmark and that suitable risk and ambiguity settings can lower the out-of-sample CVaR of dispatch costs.

Monitoring cognitive impairment (CI) in motor neuron disease (MND) is essential for timely treatment and care, yet challenging due to co-occurring speech difficulties. The Edinburgh Cognitive and Behavioural ALS Screen (ECAS) provides a robust metric for CI assessment, with the Verbal Fluency Index (VFI) a central element. Building on recent advances in automated speech analysis, this study proposes a system for estimating VFI. It leverages a unique MND dataset and combines ASR (WhisperX) and VAD (Silero) with refined timestamping to predict the VFI and extract several clinically interpretable measures. Our approach outperformed systems based on traditional acoustic features and self-supervised embeddings, evaluated using multiple regression algorithms. Clinically inspired features consistently outperformed the other sets, with the best models achieving strong results (P-words: R2 0.9, NRMSE 0.05; S-words: R2 0.8, NRMSE 0.08), demonstrating the feasibility of automated VFI estimation.

Soft continuum robots can exploit distributed compliance for whole-body manipulation, but synthesizing behavior through changing contacts remains difficult. We address whole-body grasping and pick-and-throw from an initially ungrasped state through outcome-based actuation-space optimization. Grasping is quantified by tip angular sweep and body-object enclosure, while throwing further incorporates release-direction alignment and minimum release speed. These objectives allow grasping, acceleration, and release to emerge from compliant interaction without prescribing contact forces, contact locations, or body configurations. Because the resulting actuation-to-outcome mapping is nonsmooth, we utilize derivative-free CMA-ES within an optimize-learn-refine framework. CMA-ES generates solutions for sampled conditions, a task-conditioned predictor learns warm starts, and CMA-ES refines them for unseen conditions. In simulation, the method achieves 492/500 successful grasps (98.4%) and success rates of 98%, 97%, and 94% across three directional throwing trials. Learned initialization increases grasping success from 78.6% to 98.4% while reducing the median rollout count from 1184 to 816 in CMA-ES. Hardware experiments achieve a 100% grasping success rate across 50 executions and a 100% pick-and-throw success rate across 30 executions, with 10 repetitions per direction. Together, these simulation and hardware results demonstrate the effectiveness of the proposed framework across both simulated and physical whole-body manipulation tasks.

Long-running tools can dominate coding-agent latency: compilers, test suites, and repository commands take seconds to minutes while the agent idles. This observation stall presents the same tension that drove out-of-order processors – asequential interface hides work that can be predicted and started early, but a speculative result may become visible only after it and every earlier step have been validated. We present TomasuLLM, a runtime that executes agent tool calls out of trajectory order while preserving task-execution correctness. It drafts future actions, runs them in isolated copy-on-write sandboxes, traces their dependencies and effects, and commits results in trajectory order only after validation against committed state. Across three benchmarks spanning sub-second to minutes-long tool calls, TomasuLLM improves the reported benchmark means and scales with tool latency: 1.31x on 100 SWE-bench Verified tasks, 1.35x on 28 Terminal-Bench 2.0 tasks, and 1.27x matched progress on 18 SWE-Marathon sessions. Across 4,010 audited commit-validation records, it produces zero false accepts.

The quality of a detector design is ultimately determined by the quality of the inference it enables, that is, by the accuracy with which the quantities of interest are reconstructed from the raw detector response. For complex detectors, the inference is performed by machine learning models, and the relation between the design and the attainable inference performance is, in general, non-trivial. In this work, we consider the optimization of the inference performance with respect to the detector design. The conventional approach prescribes retraining the inference model at every candidate design, thus, treating the evaluations as independent tasks and discarding the shared structure of the optimal inference algorithms at different designs. We propose the meta-learned objective estimate (MLOE): instead of solving the inference problem anew at every candidate design, a single meta-inference model, conditioned on the design and trained continually along the optimization path, is shared across all of them. We test MLOE on three families of optimization problems, the last of which comprises two design spaces of the Spectrometer Straw Tracker of the Search for Hidden Particles (SHiP) experiment; under matched budgets of simulation calls, the meta-inference model evaluates a candidate design using fewer simulation calls than the baseline strategies and holds the better rank over the convergence curve in all examined cases.

In temporal voting, a fixed set of voters makes a collective decision in each of several rounds, and proportionality requires that groups with shared interests be represented across these decisions. Full justified representation (FJR) is among the strongest proportionality axioms known to be satisfiable in this setting, since it allows different members of a group to be satisfied by different selected candidates. However, whether an FJR outcome can be computed in polynomial time has remained open. We resolve this question by introducing temporal FJR+, a strengthening of both FJR and extended justified representation+ (EJR+) that can be computed and verified in polynomial time. The axiom measures a group's agreement in each round by the largest number of its members who approve a common candidate, and requires some member to attain the integer part of the group's proportional share of its best attainable average satisfaction, with a shortfall of at most one unit when this share is an integer. We show that verification reduces to counting approvals among the voters below each satisfaction threshold. For computation, we observe that local search for Proportional Approval Voting (PAV) can stop at an outcome violating FJR. Somewhat counterintuitively, the remedy is to subtract a small multiple of the minimum voter satisfaction from the PAV score: every FJR+ violation then admits a single-round change whose improvement guarantees polynomial running time. We further show that temporal FJR+ is strictly stronger than FJR and EJR+ combined, and coincides with lower quota on party-list preferences. Finally, temporal FJR+ and multiwinner FJR+ are incomparable in general, but coincide when approvals are static and each candidate-round pair is treated as a separate candidate.

Large language models (LLMs) can give reliable labels under one setup yet change those labels when researchers make other reasonable design choices. We tested seven LLMs, 12 task designs, three independent runs, and 3,000 tweets labeled for offensive language and hate speech. Repeating the same model and task design produced high agreement (median Fleiss' $\kappa = 0.91$). Agreement fell when we changed the task design for the same tweets (median Cohen's $\kappa = 0.76$). Task design and model choice increased the variance of estimated prevalence by factors of 76.7 for offensive language and 110.6 for hate speech compared with sampling variance alone. Variation across LLM task designs reached 560-572 basis points, compared with 270-331 basis points across five human instrument versions. Confidence scores did not solve this problem. They tracked repeated model outputs more closely than agreement with human labels, and grouping six tweets in one prompt lowered mean offensive-language confidence by 660 basis points. We call the variation caused by task design and model choice instrument uncertainty. Researchers can measure it only by comparing reasonable task designs. Repeating one setup or relying on confidence scores cannot replace that test.

Vision-language models (VLMs) have made substantial progress in autonomous driving, but their success has primarily been studied in ego-centric scenes. Infrastructure-side observations provide views beyond the ego vehicle's field of view, yet conventional cooperative-driving systems typically transform them into geometric representations for downstream perception and planning. Directly incorporating these views into VLMs offers an opportunity to improve cooperative scene understanding and trajectory planning. However, question answering and trajectory planning have not been jointly evaluated on the same real-world vehicle-infrastructure scenes. We present CoVLM-Bench, a benchmark for cooperative driving question answering (CDQA) and cooperative planning (CP) on vehicle-infrastructure paired scenes. CoVLM-Bench provides scene-grounded CDQA annotations, three-part rationales as auxiliary supervision, and future trajectory targets derived from recorded ego motion. It contains 2,196 paired frames with 35,136 CDQA annotations, while CP predicts six waypoints over a three-second horizon. The annotations combine model-assisted drafting, record-based computation, and human verification. Built upon CoVLM-Bench, we introduce CoVLM-Drive, a unified VLM baseline that directly uses paired views for both CDQA and CP. Experiments show that CDQA adaptation improves answer accuracy and that CoVLM-Drive reaches a lower FDE than the compared V2X planners; QA initialization and rationale supervision each reduce planning error. Together, CoVLM-Bench and CoVLM-Drive support the training and comparison of VLMs for cooperative scene understanding and planning.

Accurate coronary artery segmentation on coronary computed tomography angiography (CCTA) is essential for diagnosing coronary artery disease. Deep networks are conventionally trained and evaluated with the Dice coefficient, but volume-overlap metrics are poorly suited to thin, tubular anatomy: since most voxels belong to a few thickproximal segments, a missing distal branch barely affects Dice despite severely disrupting the connectivity required for clinical use. We introduce a surface metric that matches predicted and reference surface points via bipartite assignment under a localized, vessel-radius tolerance, reporting precision, recall, and F1 with decoupled false positives (spurious branches) and false negatives (missed branches) a distinction the symmetric Dice cannot make. With it we show that a strong Dice-trained baseline omits far more vessel surface than it hallucinates, an asymmetry its high Dice hides. Building on this, we propose a differentiable surface loss that simultaneously suppresses spurious mass and recovers absent structure, validated by fine-tuning three backbones (nnU-Net, SwinUNETR, NexToU) on two public benchmarks (Image-CAS, ASOCA). Against a matched-epoch control, it significantly improves surface F1 by recovering missed distal vessels at comparable Dice. Our findings argue for measuring and optimizing the vessel surface, not the volume it overlaps. Code

The thermal behavior of wood is a critical factor in advanced material assembly. However, pixel-level thermal analysis remains fundamentally constrained by the low resolution and noise inherent to infrared thermography. To address this, we introduce an end-to-end computational framework that synthesizes high-resolution thermal responses directly from wood RGB images. We first establish a core physical linkage: because spatial color variation in natural wood is driven by cellular anatomy, optical intensity serves as a reliable geometric proxy for the localized solid volume fraction. By leveraging this theoretical insight, we develop an automated finite-element-method data engine that maps pixel-level optical intensity to a 3D thermodynamic voxel grid, generating high-fidelity synthetic thermal responses. We find that 1) when the thermal conductivity along the thickness direction is uniform or linear, wood RGB images and their corresponding thermal responses exhibit extreme morphological similarities, and the lateral thermal diffusion acts as a low-pass filter that smooths out high-frequency details; 2) when the thermal conductivity along the thickness direction is random, such morphological similarities are destroyed, and wood's 3D structure dominantly governs its thermal response. We further utilize these synthetic thermal responses to supervise a neural surrogate model built upon the DINOv3 foundation model. Our results demonstrate that the neural surrogate model successfully internalizes the governing thermodynamic laws, thereby bypassing computationally expensive simulations and enabling high-resolution thermal inference. This methodology effectively bridges the semantic and thermodynamic domains, unlocking systematic, pixel-level analysis of fine-grained wood thermal responses. Project: https://zekifayes.github.io/seeheat

Acute Lymphoblastic Leukemia (ALL) patients require early, accurate detection to enable timely treatment and effective patient management. A Convolutional Neural Network (CNN) is well-suited for creating an end-to-end enabling environment for ALL detection and classification. However, most CNN-based ALL detection systems are theoretical and unsuitable for deployment on edge devices due to high computational demands. The Internet of Medical Things (IoMT)-enabled devices offer an opportunity to monitor ALL patients in real time. Wearables that track temperature, heart rate, oxygen saturation, and activity can deliver critical data to support timely clinical intervention and improve patient outcomes. In smart IoMT environments, a lightweight CNN is essential because connected devices often operate under limited computational power, memory, and latency constraints. To address this need, this study proposes LukeNet, a lightweight CNN integrated with explainable artificial intelligence (XAI) for an IoMT-based SMART Acute Lymphoblastic Leukemia Detection and Management System. Trained on three (3) ALL datasets and five-fold cross-validation, LukeNet achieved an impressive 99% model accuracy as well as 99% unseen test accuracy, which is higher than six state-of-the-art (SOTA) CNNs, such as DenseNet121, MobileNet, ResNet50, InceptionV3, Xception, and VGG16, as well as transfer learning models. Furthermore, LukeNet was compared with two ensemble models. In addition, explainable AI methods are integrated to highlight relevant regions in microscopic images. The novelty of this study lies in the architecture of LukeNet, which balances model depth and computational efficiency by using depthwise separable convolutions, mitigates the risk of gradient loss in deeper layers, and provides strong global and local feature extraction capabilities.

The rise of computer vision and artificial intelligence has made possible new forms of large-scale computational measurement. We apply these techniques to a deep collection of 5,205 digitized films viewed in theaters between 1922-2025 (covering popular, prestigious, and independent movies) to trace the development of one of the most fundamental ways through which film communicates: by manipulating the space between the camera and its subject. This work finds abrupt changes with the rise of new technologies in sound and television, and allows us to shed an empirical light on gender disparity (women, despite having substantially less screentime than men, are disproportionately the subject of closer shots), and illustrate how animated films both inherit and break free from the norms of live-action filmmaking.

Retrieval-augmented generation is an attractive way to improve chest X-ray reporting, because reports from similar prior studies supply clinical context a general vision-language model lacks. We show the same mechanism is a reliable source of clinical error. Over 100 MIMIC-CXR studies with a frozen LLaVA-1.5 generator and BioMedCLIP retrieval, CheXbert clinical F1 doubles under relevant retrieval (0.201 to 0.402) and collapses to 0.043 under clinically mismatched retrieval, a fifth of the image-only score; the retrieval-induced hallucination rate, counting only unsupported findings traceable to retrieved evidence, rises from 0.00 to 0.76 and 0.98. To show this is not an artefact of evidence-set coverage, we introduce a coincidental-overlap control that scores image-only generations against evidence they never saw, placing the chance base rate at 0.18, four to five times below the observed rates. The generator does not merely acquire findings, it transcribes text: 95% of reports produced under relevant retrieval contain an eight-word span occurring verbatim in the retrieved evidence but absent from the reference, against 0% without retrieval. We then explain the mechanism: a normal chest X-ray retrieves at least one abnormal precedent in 24 of 28 cases, because medical image-embedding similarity is dominated by anatomy and acquisition rather than by the presence of disease. This has a direct design consequence. Retrieval similarity does not predict harm (RIH rates 0.80/0.80/0.60/0.84 across similarity quartiles), so relevance gates conditioned on embedding similarity cannot work; gating on predicted pathology agreement removes 61% of unsupported evidence at no cost to useful coverage. We argue that retrieval-augmented clinical systems must be evaluated under retrieval failure, not only under retrieval success.