Papers

Industrial procurement requires language models to bridge informal buyer jargon, sparse marketplace attributes, and authoritative engineering standards under strict safety tolerances. We present IndustryLLM, an open-weight industrial language model trained from Qwen3.5-35B-A3B-Base (35B total parameters with ~3B activated per token, with the vision encoder frozen). Rather than relying on generic text scaling, we introduce a failure-driven adaptation recipe spanning continued pre-training (CPT) and supervised fine-tuning (SFT). CPT leverages a curated ~100B-token corpus integrating 5B tokens of national standards (e.g., GB/T) and technical archives, 10B tokens of de-identified real-world industrial transaction and inquiry records, and 60B tokens of general replay. To overcome register mismatch and factual brittleness, we systematically reconstruct an estimated 20B-token domain subset via multi-register rewriting across 10 genres and 8 writing styles, confidence-routed minimal factual editing, and error-targeted QA synthesis (resolving colloquial typos like '42-luo-mu' -> 42CrMo, expanding ambiguous codes like '16674' -> GB/T 16674, and clarifying conflicting dimensional specs). For downstream deployment, we formalize an evidence-gated constraint-evaluation interface enforcing three-valued logic where unverified product evidence remains unknown rather than satisfied. Offline evaluations demonstrate consistent gains on procurement-query structuring (+2.97 percentage points in exact match, 95% CI [2.11, 3.86] in No-Think mode), while randomized online A/B experiments in production yield substantial improvements (+4.25% GMV, +8.3% satisfied inquiries) alongside a latency reduction from 6-7 s to 1.5 s. Model weights and configs are released at https://huggingface.co/alibaba-multimodal-industrial-ai/IndustryLLM.

Consistently profitable trading is difficult because equity markets are noisy, non-stationary, and only partially predictable from historical data. We benchmark five deep reinforcement learning (DRL) actor-critic methods: A2C, PPO, DDPG, TD3, and SAC, that learn trading actions end-to-end from market states, and compare them with a supervised price-forecasting baseline. Using daily data for 20 large-capitalization S&P 500 stocks from 2000 to 2020, enriched with trend-following technical indicators and log min-max scaling, we train on 2000-2018 and backtest on 2019-2020. Each agent is evaluated both when trained once and under forward retraining, in which it is retrained on all data available before each successive test window. DDPG achieves the highest annual return (55.5%), Sharpe ratio (1.38), and alpha (0.22), but also the highest market beta (1.24). TD3 and SAC offer a better risk-return balance, with Sharpe ratios of 1.37 and 1.33 and maximum drawdowns of about 25%. Forward retraining improves A2C, PPO, and SAC, leaves TD3 essentially unchanged, and reduces DDPG's annual return from 55.5% to 29.8%, consistent with TD3's greater robustness to hyperparameters. The forecasting baseline has the smallest maximum drawdown (9.6%) and the lowest beta (0.31), underscoring a trade-off between the higher returns of end-to-end DRL and the lower risk of forecast-driven strategies.

Open-vocabulary keyword spotting (KWS) must detect arbitrary keywords without retraining. Cross-attention-based models achieve state-of-the-art performance but require pairwise interaction between the audio and each keyword, recomputing the fused representation for every audio–keyword pair. Embedding-based models avoid this computational cost through independent encoding and similarity scoring, yet remain inferior on challenging benchmarks such as LibriPhrase-hard. We show that this gap is primarily a training issue rather than a limitation of model capacity. While contrastive learning encourages correct keywords to rank above negatives, it does not explicitly calibrate absolute similarity scores, which are critical for applying a fixed threshold across keyword detection. We propose Calibrated, Order-aware Detection KWS (CORD-KWS), a training framework that introduces two complementary objectives while retaining the same encoders, cosine scoring, and O(1) keyword enrollment: a calibrated detection head that supervises absolute scores, and a frame-level CTC objective that preserves phonetic order information. CORD-KWS achieves EERs of 0.43% and 9.64% on LibriPhrase-easy and LibriPhrase-hard, respectively, outperforming existing embedding- and cross-attention-based models.

Model-predictive control with Joint-Embedding Predictive Architectures (JEPAs) provides a strong zero-shot goal-reaching planner, but it is only effective over short planning horizons. Hierarchical extensions attempt to bridge this gap by learning a macro planner to predict intermediate latent sub-goals to guide the micro planner. In this work, we demonstrate that unconstrained latent sub-goal prediction is fundamentally flawed. A rigorous evaluation reveals that leading hierarchical macro planners routinely emit physically unrealisable sub-goals. To resolve this, we introduce Metro-WM, a hierarchical framework that issues sub-goals by retrieving genuine states from prior experience rather than generating ungrounded latent vectors. Specifically, Metro-WM constructs a graph whose vertices are observed frames from offline expert demonstrations or random-action trajectories, allowing frames from different episodes to be connected and stitched into routes to the goal. Planning over the full graph also makes the system highly robust to execution errors: if the micro planner drifts off course, Metro-WM instantly finds a new optimal path from the current state. Our experiments show that Metro-WM achieves superior long-horizon success rates of up to 37.33 percentage points over the strongest hierarchical baseline on the same world model while being up to 10.9 times faster, requiring both 13-56 times less offline compute and fewer tuned hyperparameters. Additional analysis reveals that Metro-WM frequently finds shorter paths than the offline demonstrations, outperforms an oracle relying on the query's own demonstration on most benchmarks, and maintains robust performance under extremely sparse dataset conditions.

This paper presents AirLog, a smartphone-based life journaling system that automatically reconstructs users' store visits in shopping malls and summarizes them into human-readable journals. Unlike conventional indoor localization systems, AirLog avoids labor-intensive radio-map construction and dedicated wireless localization infrastructure and algorithm calibrations. Instead, it repurposes two cues already available in commercial spaces: semantic information exposed by ambient Wi-Fi SSIDs and indoor directory images. AirLog converts directory images into spatial maps and fuses Wi-Fi semantic anchors with inertial dead reckoning to recover store-level trajectories, which are then summarized into journals by an LLM. Such store-level life logs can support applications such as personal memory recall, activity reflection, and automated diary generation without requiring users to manually record where they have been. We implement AirLog on commodity smartphones and evaluate it on both a large-scale public dataset and a self-collected dataset. The results demonstrate that AirLog substantially improves store-level region recovery, semantic matching, trajectory reconstruction, and journal quality over existing baselines. A human evaluation further shows that the generated journals are coherent and faithful to users' visits.

Reconstruction and simulation at a collider experiment are long chains of specialised algorithms, each tuned to a single step. We explore how one model can serve many of those steps at once, while still producing the intermediate objects (tracks, calorimeter cells, clusters, particles and jets) that make the chain interpretable. To do so, we represent every object in a jet in a single shared token vocabulary and train one model to map any subset of these modalities to any other. A task is then only a choice of which modalities to provide and which to request: particle flow, detector simulation and charged energy subtraction are all directions through the same set of weights. We train over all modality combinations, with a decoder emitting tokens either autoregressively or in parallel. With tokenisation, both architectures train stably with little tuning. In evaluations, both models produce realistic reconstruction and simulation objects, with the autoregressive model particularly faithful to output from Geant4. The autoregressive model also outperforms a state-of-the-art particle flow algorithm on many typical jet reconstruction metrics.

Validating an LLM-as-a-judge requires estimating its agreement with humans, yet annotation budgets rarely allow every item to be multiply labeled. We prove that this overlap sparsity is the first-order determinant of wrong deployment decisions: at 5% pairwise overlap, wrong-decision rates reach 25% and the probability of selecting the wrong best judge among ten candidates is 65%. The two actionable levers are overlap quantity and allocation. For quantity, we derive a minimum-overlap formula showing $\rho \geq 0.25$ suffices for non-borderline judges while borderline cases remain fundamentally hard. For allocation, a zero-cost stratified scheme halves false-rejection rates relative to random sampling when strata are informative. We validate on 10 LLM judges across four evaluation matrices spanning visual assessment, causal reasoning, and summarization.

Human demonstrations provide dense policy-level information but sometimes lack local precision. Human feedback presents accurate local critiques, but offers sparse evaluations rather than direct policy guidance. We propose Progress-Heuristicized Inverse Reinforcement Learning (PHIRL), a data-efficient framework that learns robust reward functions by jointly leveraging demonstrations and feedback. Specifically, we use progress, a feedback modality that describes cumulative task completion. PHIRL iteratively infers a reward function from demonstrations via inverse reinforcement learning, calculates the learned rewards over the progress-annotated demonstrations, and aligns the rewards with progress annotations over four dimensions. We evaluate PHIRL on real and simulated robot tasks, with additional exploration using a fine-tuned vision-language model to provide progress feedback. Results demonstrate that PHIRL significantly outperforms the baselines, achieving substantially higher environmental return rewards and task success with only twenty percent of demonstrations annotated. Analysis of reward-hacking scenarios demonstrates that PHIRL learned reward functions are reliable against exploitation.

Rational stability is a quantitative stability notion for nonlinear ordinary differential equations (ODEs), under which convergence is characterized by a rationally-in-time decaying bound on solutions. Although Lyapunov characterizations of rational stability have been proposed, no existing condition has been shown to exactly characterize rational rate performance. The paper resolves this issue by providing an exact converse Lyapunov characterization of regional rational stability with prescribed rate performance. To test the resulting converse Lyapunov conditions numerically, a tiered set of convex relaxations is proposed which can be enforced using Sum-of-Squares (SOS) programming. A conjectured conservatism scaling factor in rate performance is associated with each tier and validated numerically. The first proposed tier appears to be novel and is shown to outperform classical SOS-based approaches. Numerical examples are used to validate the results and compute nested regions of state-space on which prescribed levels of performance can be guaranteed.

The IceCube Neutrino Observatory has removed GridFTP and x.509 certificate authentication for data transfers, migrating to the Pelican Platform, the Open Science Data Federation, and WLGC tokens. While this is a common solution on the computing infrastructure we use, we required several customizations to work with our existing data storage structure and make it easier for scientists to use. We wrote a custom WLCG token issuer to support our POSIX filesystem with custom user and group permissions across the entire filesystem. We also made several modifications to HTCondor to ease job submission using tokens, including a custom credmon. After an initially bumpy transition due to several now-resolved Pelican issues, the Pelican-based system has already proven superior to GridFTP in several ways.

We consider the fair division problem with indivisible goods and study the following decision problem: given a fair division instance, does there exist an allocation that is envy-free and efficient? We consider two efficiency criteria: Pareto-optimality and social welfare optimality. We provide a complete landscape on the computational complexity of this decision problem, with the number of agents ranging from $2$ to $\infty$, both additive valuations and general valuations, and the more restricted class of $k$-ary valuation functions (where an item's marginal value is restricted to $\{0,1,\ldots,k-1\}$ for some constant $k\geq2$). One interesting observation is that many versions of the above-mentioned decision problems fall into the “gap” between the first and the second levels of the polynomial hierarchy. Specifically, assuming the polynomial hierarchy does not collapse to the first level (i.e., assuming $\text{NP}\neq\text{coNP}$), these problems are in $(\Sigma_2^{\text{p}}\cap\Pi_2^{\text{p}})\setminus(\text{NP}\cup\text{coNP})$. In particular, we provide a fine-grained complexity analysis across different parameter regimes, including the number of agents and the choice of valuation models. Depending on different parameters, many problems admit different complexity classifications, ranging from the intermediate classes $\Theta_2^{\text{p}}$ and $\Delta_2^{\text{p}}$ between the two levels to $\Sigma_2^{\text{p}}$-completeness. Finally, De Keijzer et al. show the $\Sigma_2^{\text{p}}$-completeness of the decision problem when considering Pareto-optimality as the efficiency criterion with additive valuations. Our main results extend this result to more restricted settings, such as instances with a constant number of agents or $3$-ary valuation functions, which resolves the open problem given by Bouveret and Lang.

The Gromov-Wasserstein (GW) distance measures the discrepancy between metric measure (mm) spaces and identifies optimal alignments between them based solely on their intrinsic structure. Since it identifies isomorphic mm spaces, it provides a natural notion of distance for heterogeneous datasets which may admit isomorphic representations. In order to accelerate computation of GW distances, many practitioners employ entropic regularization to obtain an Entropic GW (EGW) problem. The most popular EGW solver is the Mirror Descent (MD) algorithm, which reduces EGW computations to an iterative process where an entropic optimal transport (EOT) problem is solved at each iteration. Despite its widespread use, the convergence of MD for this problem has only been established for restricted classes of costs. On the other hand, a recently proposed dual gradient method is available for general costs, but requires a choice of step size which depends on the regularization parameter. To address these two issues, we introduce Averaged Mirror Descent (AMD), which averages consecutive MD steps, and prove its convergence for arbitrary costs. Then, we establish that the dual gradient method with a fixed step size also converges for arbitrary costs at the cost of a more complicated iteration. In both cases, we also account for inexact iterations which are inescapable in practice. We compare the empirical performance of these methods across various settings and, in particular, show that AMD and the dual gradient method both converge on an example where classical MD fails.

General-purpose agents can plan, reason, and act over long horizons, yet their production capabilities remain fragmented across text, images, audio, video, documents, 3D assets, and code. Extending a foundation model to additional modalities ties capability growth to costly model updates, while assembling specialist models and tools leaves unresolved how procedures, dependencies, intermediate assets, and cross-turn revisions should be coordinated. We present Omni-IO Skills, a plug-and-play Agent Harness that makes existing agents omni-native through hierarchical Skills, a standardized multimodal execution interface, dependency-aware orchestration, and a persistent Asset Registry. Multi-asset workflows are represented as Declare Execution Graphs, which schedule independent operations concurrently and register successful outputs for downstream and cross-turn reuse across replaceable execution backends. Its 27 Skills cover 38 representative tasks spanning seven artifact modalities and four capability families: understanding, generation, reasoning, and retrieval. On UniM-90, the harness raises the input-support rates of GPT-5.6 Sol and Claude Sonnet 5 from 40.00% and 38.89% to 100%, while increasing relative Semantic–Quality Coupled Score from 26.99 to 74.94 and from 27.82 to 77.78, respectively; Strict Structure Score reaches 100.00 and 99.78. These results establish harness-level capability composition as a practical route to broad, evolvable Omni systems without changing the host agent's reasoning core.

High-speed racket sports provide a demanding testbed for humanoid robots, requiring time-critical decisions, precise striking, and dynamic whole-body coordination. In badminton, fast-changing shuttle trajectories require timely contact decisions, while successful returns demand precise racket pose and velocity within a brief contact window and across a broad three-dimensional striking workspace. Human motion data provide valuable priors for such athletic skills, but usable badminton references are limited and imperfect. Direct tracking provides insufficient executable variation for diverse shuttle conditions, while purely task-driven optimization may produce unnatural motion. We therefore present a three-stage hierarchical reinforcement learning framework for dynamic humanoid badminton. First, task-randomized motion augmentation expands sparse annotated hitting events into executable target-conditioned stroke variations, forming a continuous latent skill space. Second, a high-level planner outputs continuous latent skill codes to compose these skills online according to the observed shuttle state. Third, a context-conditioned adversarial regularizer encourages more natural planner-level skill usage while preserving return performance. In real-world deployment, our system sustains multi-skill rallies with human players, including forehand, backhand, and highly dynamic jump returns. This is the first real-world humanoid racket-sport system to demonstrate multi-skill human–robot rallies including highly dynamic jump returns.

A confluence of recent works has shown that many quantum circuits and dynamics are efficiently simulable by classical algorithms that track local information, even when conventional complexity measures such as the entanglement and magic are high. Here, we introduce a novel measure of complexity, the reactivity, to capture this new method of classical attack. Unlike conventional complexity measures, the reactivity does not capture a property of a quantum state or operator in isolation, but rather a quantum experiment as a whole. We provide numerical and rigorous evidence that quantum experiments with low reactivity are simple by a host of measures: they are efficient to classically simulate, learn, and fast-forward. This motivates the search for quantum experiments with high reactivity, which may evade these simplistic features. To this end, we introduce easily implementable experimental protocols—dubbed Pauli path spectroscopy—that allow one to efficiently measure the reactivity of any quantum experiment of interest. Our protocols are applicable even when the experiment itself is beyond the reach of classical simulation.

Large vision-language models (LVLMs) have demonstrated remarkable performance on multimodal reasoning benchmarks, yet their perceptual reliability under physically constrained imaging conditions remains poorly understood. Existing evaluations predominantly assume ideal visual inputs and therefore fail to characterize how camera distance, illumination, viewpoint, and pixel density fundamentally affect semantic recoverability. We introduce SynDORBench, the first physically grounded benchmark for evaluating LVLM perceptual robustness under DORI-calibrated conditions aligned with human visual capability standards. SynDORBench comprises over 54k question–answer pairs generated through a controllable synthetic pipeline that systematically varies viewing distance, lighting, camera geometry, and action pose according to physically interpretable pixel-density regimes. To support scalable low-visibility supervision, we further propose a discernibility annotation framework that propagates human perceptual labels using mask-conditioned statistical features and ensemble learning. We evaluate 16 open-source LVLMs, a commercial LVLM baseline, and YOLO11x across human-presence classification and action recognition tasks under progressively degraded visibility conditions. Our results reveal that perceptual failure in LVLMs is strongly governed by pixel density and physical imaging constraints rather than model scale alone. Surprisingly, several compact open-source LVLMs outperform larger commercial baselines and substantially exceed YOLO11x robustness under long-range and low-light conditions. SynDORBench establishes a new benchmark paradigm for physically grounded multimodal evaluation, enabling systematic analysis of LVLM reliability under real-world perceptual constraints and direct comparison against human visibility thresholds.

Developing label-efficient models is a central challenge in surgical AI due to the high cost and scarcity of expert annotation. While self-supervised foundation models adapt well to new tasks with minimal data, how label efficiency varies across different surgical tasks remains largely unexplored. Here, we introduce SURGE, a surgical foundation model trained on SurgSpectrum-30M+, the largest pretraining dataset comprising over 30 million frames, with checkpoints released to enable further research. We systematically evaluate label efficiency across 5 task categories and 15 benchmarks. These range from temporal and spatial scene understanding to fine-grained reasoning tied to instrument-anatomy interactions and safety-critical maneuvers. SURGE outperforms prior state-of-the-art on all benchmarks, even surpassing task-specific models on complex reasoning tasks. Crucially, we reveal a task-dependent scaling behavior: while scene understanding tasks saturate with minimal supervision, fine-grained reasoning tasks continue improving with substantially larger annotation budgets, providing a blueprint for allocating expert effort in complex domains. Code: https://github.com/CAMMA-public/SURGE

Hypothesis-awkward is a collection of Hypothesis strategies for Awkward Array. Awkward Array can represent a wide variety of nested, variable-length, mixed-type data. Many tools that process Awkward Arrays are widely used and actively developed. The unit test cases of many of these tools list predefined input samples. In practice, such samples can cover only a small portion of the vast combinatorial space of Awkward Array instances. Hypothesis, a Python property-based testing library, generates test data that can make test cases fail and automatically explores edge cases. The main strategy of Hypothesis-awkward generates nearly all possible Awkward Arrays, with options to control the layout, data types, missing values, masks, and other array attributes. Property-based tests written with these strategies have begun to run in the continuous integration of Awkward Array; to date, they cover a small part of Awkward Array and have found 37 bugs in it, as well as one in Hypothesis itself. These strategies make the tests of Awkward Array and of the tools that use it more reliable.

The Koopman operator is a well-established framework for lifting nonlinear dynamical systems to an infinite-dimensional space where dynamics are linear. Extended Dynamic Mode Decomposition (EDMD) is a widely used method in data-driven dynamics as it provides a Galerkin approximation of the Koopman operator on the finite-dimensional span of a prescribed dictionary of observable functions. In particular, EDMD only requires samples from the dynamical system, describing the Koopman operator without knowledge of the underlying dynamics. While many studies analyze asymptotic convergence results for EDMD in terms of data, the question of more practical finite-data convergence results remains incomplete in general settings. In this work, we prove bounds for the finite-data EDMD Galerkin approximation of the Koopman operator by a rate proportional to the inverse of the root of the number of samples (the Monte Carlo rate). These results apply to multiple classes of problems, i.e., for discrete or continuous dynamics, including stochastic systems. In certain more structured setting, we demonstrate super-polynomial rates are achievable both theoretically and computationally. Finally, we derive recovery guarantees for an EDMD variant in the undersampled regime inspired by compressed sensing techniques.

What should a compressed representation preserve when the information of interest lies in relationships among elements rather than in the elements themselves? We formulate relational compression in the classical source-description-reconstruction sense, but with relational structure itself as the fidelity-bearing content. Each instance specifies the source relation, retained description, reconstructed or evaluated relation, fidelity criterion, and constrained resource. We use this interface to situate selected methods from graph summarization, spectral sparsification, similarity-preserving representation, and relational distillation within a common formulation while keeping their different reconstruction and resource assumptions explicit. We develop finite-codeword collision as one concrete realization. Same-codeword probability yields a relational geometry linking pair-specific alignment and separation to aggregate Rényi-2 occupancy and the spherical geometry of categorical assignments, with exact objective correspondences to squared-Euclidean centroid reconstruction and normalized graph association and cut. Graph and image studies illustrate complementary routes within the finite-codeword family: graph- and teacher-defined relational requirements act directly on equality or collision, while reconstruction acts through a joint decoder. Together, these results illustrate how distinct relational requirements can be formulated and tested within a common constrained-representation framework.

Spreadsheet disasters have been documented for more than thirty years, but they have not gone away. This paper examines the EuSpRIG horror stories archive as evidence of how ordinary spreadsheets fail and the consequences that follow. After cleaning and deduplication, 121 incidents from 1995 to 2026 were classified into twelve failure categories and analysed by mechanism, context and consequence. Formula errors, data-entry slips and data-handling mistakes account for 71 cases and recur across the whole period, from Fidelity's missing minus sign in 1995 to a mistyped date that cost Norway's sovereign wealth fund about $92 million in 2024. A second pattern concerns information present in a workbook but not visible to the recipient: hidden rows, hidden sheets, pivot-table caches, embedded objects and retained source layers can cause harm even when no calculation is wrong. The clearest rise in reported cases is in this hidden or embedded data category, whose most serious example is the 2022 Ministry of Defence spreadsheet that exposed details of roughly 18,700 Afghan applicants through hidden rows. The paper argues that spreadsheet risk is no longer only a problem of wrong numbers. It also involves visibility, disclosure, governance and trust. Existing taxonomies remain useful for formula, entry and handling errors, but do not fully capture the public harm caused by spreadsheets used as uncontrolled operational infrastructure. The same mistakes have persisted for three decades, what has changed is the authority organisations give to the spreadsheets that contain them.

Evaluating VLM-based autonomous driving remains difficult because driving competence is composite, where a capable system must ground traffic participants and hazards, integrate context across views and time, reason about future evolution, and act appropriately under closed-loop interaction. Existing benchmarks usually assess either open-loop understanding or closed-loop driving but provide limited structure for explaining how these abilities are organized, how they relate, and how they may inform model diagnosis and improvement. We present DriveHierarchy, a hierarchical benchmark that organizes VLM-based autonomous driving into four ranks, spanning perceptual grounding, contextual memory, mental reasoning, and closed-loop execution. To instantiate this hierarchy, we integrate multiple open-source autonomous-driving datasets into a unified open-loop benchmark with 76,798 question-answer pairs over 84,279 frames and develop a closed-loop simulation platform with interactive scenario construction on a real-world road network, from which 100 driving scenarios are curated for embodied evaluation. Experiments on 15 VLMs show that DriveHierarchy captures structured but non-redundant capability variation, relates open-loop understanding to closed-loop driving, and provides a practical basis for diagnosis and benchmark-guided optimization. DriveHierarchy therefore serves as a unified framework for evaluating and improving VLM-based autonomous driving systems. An anonymized project has been released on https://github.com/PerfectXu88/DriveHierarchy

Dormant Awakening (DA) specializes a magnetic resonance imaging (MRI) reconstruction model by fitting a small set of zero weights to one labeled slice. Across fourteen U-shaped convolutional network (U-Net) and vision transformer (ViT) sources, fitting 1,004 or 10,560 weights gives mean peak signal-to-noise ratio (PSNR) gains of .430 and .509 dB. We analyze how scaling the fitted output correction changes evaluation PSNR, then use the calibration correction size to cap changes on new images. The cap bounds changes in root mean squared error (RMSE) relative to the source without an evaluation reference. At 10% adaptation budget, we compare fixed halving and dynamic capping for DA, unrestricted sparse adaptation and low-rank adaptation (LoRA). The controls increase both measured calibration-evaluation correlations across all six tested architecture-adapter settings. Experiments concern one constructed fastMRI-to-M4Raw shift with fixed source checkpoints and participant panels.

Pruning reduces the number of weights in magnetic resonance imaging (MRI) reconstruction networks. Equal sparsity, however, can retain weights with different computational roles and different compatibility with the trained network. We study these effects across 120 convolutional U-Net and vision transformer models using controlled edits evaluated before retraining. In U-Net, preserving high-resolution computation improves reconstruction at equal deletion counts across all tested sparsity levels. Equal-operation controls reveal additional sensitivity of the first convolution, which operation count alone cannot explain. In transformers, retained pretrained weight energy correlates with quality within sparsity levels, yet the same masks reverse their reconstruction ranking when the surrounding trained weights change. At 90% sparsity, matching intermediate output scales largely removes a high-energy-mask penalty while retaining an advantage for the network's original mask. Thus sparsity alone does not characterize reconstruction quality: architecture-specific descriptors are informative, but mask quality can still depend on the surrounding trained network.

Video-to-music models have advanced considerably in the last few years, particularly in film and music video applications. In this paper, we investigate this problem in the video game domain, which introduces new challenges for these models: video frames are rendered graphics, music is mostly synthetic audio, and soundtracks loop across entire levels rather than following on-screen events. We introduce a new dataset of 217.6 hours of Super Nintendo (SNES) gameplay video paired with 485 hours of clean soundtracks, free of sound effects and voice-overs, matched to gameplay audio via audio fingerprinting. With this dataset, we train a simple encoder-decoder transformer that passes video features directly to a MusicGen decoder, comparing different encoding strategies: textual descriptions (T5), independent frames (ViT), or spatiotemporal patches (ViViT). Each encoder is tested both frozen and fine-tuned, while the decoder is always fine-tuned. Frozen encoders match or outperform their fine-tuned counterparts on every metric, and the frozen ViViT achieves the best overall results. We compare this model with state-of-the-art baselines using both objective metrics and a listening study (N = 96). Despite having up to 18% fewer parameters, our model outperforms all baselines on objective metrics, surpasses GVMGen in the listening study, and performs comparably to OSSL.

Accurate medium-term, from a few months to a few years, electricity load forecasts are crucial for informed decision-making in power plant maintenance scheduling, load dispatch and price settlement. Being comprised between Long-Term Load Forecasting (LTLF) which uses mostly economic projections and appliances development scenarios, and Short-Term Load Forecasting (STLF) driven by weather, calendar and autoregressive patterns, Medium-Term Load Forecasting (MTLF) requires both extrapolation capabilities and variability modeling. Yet, it remains unclear if MTLF can benefit from economic indicators, and especially at which forecast horizon and resolution. To address these challenges we investigated the impact of socioeconomic data on predictions issued 1 month and up to 48 months in advance for France at monthly and daily resolution using a tabular Foundation Model (FM). A dataset covering 20 years of observations of electricity load, weather variables and economic features such as consumer price and production indices, electric vehicle counts or employment is created for the study. To avoid noisy data, we used a new feature selection pipeline, creating ensemble of expert models with diverse feature subsets, to demonstrate that selected economic covariates improve forecast skill by 20% over 2015-2025. This enhancement is steady across lead times and resolutions limiting the Mean Absolute Percentage Error to 4% for monthly granularity and 5% for daily granularity. Explainability of the models is investigated through feature and context importance. Results showed that the FM is limited in the context it leverages pointing towards potential computational savings with a reduced context, while feature importance of economic predictors grows with the forecast horizon. This suggests that including economic data in MTLF could bridge the gap with LTLF leading to seamless forecasts.

Space weather disturbances driven by the solar wind can degrade satellite navigation, disrupt radio communications, and threaten power infrastructure, making accurate forecasting of the high-latitude ionosphere's response a critical operational need. Existing approaches, empirical climatological models and physics-based magnetohydrodynamic simulations, either smooth out the ionosphere's time-dependent, non-linear response or are too computationally expensive for real-time use, while prior machine learning efforts have mostly targeted scalar indices rather than the full spatial structure of ionospheric convection. Here we train and compare three deep learning architectures, a probabilistic diffusion model conditioned on multi-variate solar wind and interplanetary magnetic field measurements at the L1 Lagrange point, an unconditioned diffusion ablation, and a deterministic U-Net baseline, to forecast Southern Hemisphere high-latitude electric potential maps derived from SuperDARN radar observations. Using five years (2020–2025) of SuperDARN data synchronised with DSCOVR L1 measurements, we evaluate the models under a single-pass regime, an extended autoregressive rollout of up to 350 frames, and an out-of-distribution case study on the intense March 2015 St.\ Patrick's Day storm, unseen during training. We find that the relative advantage of the deterministic and probabilistic approaches is not fixed: the deterministic model is competitive over short horizons and calm conditions, while the diffusion model's advantage grows and eventually dominates as the forecast horizon lengthens and the event becomes more dynamic, evidence that probabilistic, sample-based generative models are the more promising direction for operational, long-horizon space weather forecasting.

Autonomous navigation in dynamic environments is a critical challenge, particularly when spaces are shared with other mobile agents whose future trajectories are known. While traditional grid-based planners efficiently find collision-free paths, their reliance on stop-and-turn mechanics over $2^k$-connected grids produces piecewise-linear trajectories that are kinodynamically highly sub-optimal for differentially constrained robots. In this paper, we present an adaptation of Safe Interval Path Planning (SIPP) that operates on state lattices, utilizing precomputed, kinodynamically smooth motion primitives. To efficiently handle dynamic environments, we rasterize the spatiotemporal swept volumes of moving obstacles directly onto the grid, treating grid cells as atomic units of space, whose resolution is typically dictated by inherent localization noise. We perform a comprehensive comparative analysis between our lattice-based approach and $2^k$-connected grid planners across diverse topological environments. Our evaluation considers a broad spectrum of performance metrics, including planning time, path angularity, cumulative heading change (angle-over-length), and bending energy. The results demonstrate that while the expanded state space of lattice-based search increases computational overhead, it yields trajectories with significantly superior kinodynamic properties. Specifically, our method achieves a reachability comparable to highly connected grids while ensuring smooth, continuous, and physically executable paths ready for real-world deployment.

Complex systems such as vehicles, patients, or genomes emit discrete event sequences whose operative question is causal, not predictive: which events cause which other events, and which cause higher-level outcomes such as failures or diseases? This question decomposes along two axes – dependency type (event $\to$ event vs.\ event $\to$ outcome) and causal scope (single sequence vs.\ population) – yielding four structurally distinct regimes with different identifiability conditions. No existing method addresses more than one, because all assume multi-stream structure with low vocabulary, and none scales beyond a few hundred event types. We present Seq2Cause, a unified framework that resolves all four regimes through a single shared primitive: a pretrained autoregressive model repurposed as an amortized conditional independence testing engine requiring no task-specific retraining. We establish a prediction–causality duality: the model's excess cross-entropy simultaneously bounds causal identification error across all four regimes, so that every improvement in next-token prediction tightens causal guarantees for free. On nonlinear SCMs (vocabularies up to $8{,}000$ types) and real-world vehicle diagnostic logs ($29$K event types, $474$ failure outcomes), Seq2Cause is the first method to populate all four regimes at scale with a single frozen backbone. Existing methods are either inapplicable, inaccurate, or computationally intractable in this setting.

We prove dimension-independent stability for the Courtade-Kumar inequality: a Boolean function $f:\{-1,1\}^n\to\{-1,1\}$ whose information is close to the dictator value is close in probability to a signed dictator. The correlation dependence is sharp in order near zero and, for increasing functions, also at the noiseless endpoint.