让下一篇重要论文，不再错过。

Anomaly detection (AD) is often framed as a representation-learning problem, where strong performance is expected to require learned features or AD-specific optimization. However, this view can overlook the intrinsic separability of anomalies from normal references, even in the absence of learned representations. We therefore investigate how far untrained neural networks can go in anomaly detection. To this end, we propose PRISM, a framework combining fixed random nonlinear features with normal-reference Mahalanobis scoring for tabular detection and dense visual localization. Across 47 tabular datasets in ADBench, PRISM-T achieves the highest mean AUROC (86.4%) and the second-highest mean AUPRC (65.9%) among the compared methods. On MVTec AD and VisA, covering 27 visual categories, PRISM-V achieves mean image-level and pixel-level AUROCs of 90.4% and 96.4%, respectively. Notably, both models achieve these results without neural weight optimization or pretrained representations. A theoretical analysis further characterizes how nonlinear transformations affect the mean and covariance contributions to the Mahalanobis score gap, while empirical analyses examine the resulting score separability. These findings establish PRISM as both a competitive anomaly detector and a prior-to-learning baseline for assessing the added value of learned representations in anomaly detection.

AI evaluation is increasingly central to AI safety and governance, yet taking part in an evaluation does not by itself confer authority over its design, its execution or the resources it depends on. This paper examines how infrastructure arrangements distribute that authority, and so shape who can evaluate AI systems independently. Drawing on Science and Technology Studies, political economy and applied evaluation standards, we introduce PAAI (Power, Access, Adequacy, Independence), a framework for examining who controls evaluation infrastructure, the terms on which it is available, whether it is adequate to the evaluation's purpose and what autonomy it supports. We apply PAAI to twelve African AI evaluation projects (2022-2025) and to the documented roles of all ten member institutes in the three published joint testing exercises of the international network of AI Safety Institutes. We identify three recurring configurations: extractive collaboration, dependent innovation and constrained independence. African actors often contribute data, language expertise and task design while external institutions retain authority over infrastructure and evaluation frameworks; design authority without infrastructure authority never yielded more than partial independence. The joint testing exercises show a similar division among mostly high-income members: the institutes that led each exercise also ran the models, while Kenya, the network's only African member, moved from annotation to task design with no documented model execution. Participation in evaluation does not by itself produce evaluative authority. For evaluation to serve as a credible governance mechanism, access must come with control over priorities, methods, publication and continuity.

Production deployments often use different models to balance cost and quality through cascading, routing, and switching models during long conversations. Context accumulates across turns. Each switch requires the target to repeat prefill over this context. Existing methods transfer full-attention (FA) K/V caches across models. Hybrid linear-attention models also retain recurrent states (REC) and convolution histories (Conv). The interaction between Conv and REC complicates cross-model state transfer. Conv perturbations can change keys and values in recurrent layers and alter REC updates. Conv histories are overwritten, but changes in REC can still affect later outputs. More accurate REC reconstruction need not improve continuation. Building on this, we propose Hybrid State Transfer (HST) for all three state types. It uses structured REC maps and preserves history order during Conv alignment. We calibrate the adapters using the frozen target's outputs during continuation. FA cache map updates are merged offline. Across five benchmarks, Qwen3.5 4B$\rightarrow$27B and 9B$\rightarrow$27B achieve mean task retentions of 89.88% and 94.11%, respectively. With source caches already available, separate timing on synthetic 16K inputs yields handoff body speedups of $21.66\times$ and $21.91\times$.

Predictive interference management supports ultra-reliable low-latency communication (URLLC) by allocating finite-blocklength resources against a high conditional quantile of next-slot interference. Building on an extreme-value-theory/kernel-density-estimation (EVT/KDE) predictor with quantile-defined interference states, this paper makes the predictor resilient to unannounced shifts in the interference distribution. Its cumulative refitting on all past observations does not specify how the state partition and retained samples should evolve after a regime change. We show that drift first makes the quantile partition inefficient: cumulative refitting preserves tail support and coverage but cannot recover conditional resolution, and pays for it in channel use, whereas discarding history starves the tail estimator. We therefore propose event-triggered quantile-space adaptation. A nonparametric cumulative sum (CUSUM) on the predictor's level stream detects change. At each confirmed event, the quantile partition is re-derived from post-change observations, retained history is re-binned under it, and each state is refit from a bounded per-state memory, capped to balance stale-regime bias against tail-estimation variance, with a closed-form L-moment tail estimator. At a nominal coverage target $η\!=\!0.99$, relative to matched cumulative refitting with fixed states, the proposed method reduces channel use by $5.7$--$13.2\%$ and quantile pinball loss by $25$--$50\%$ across abrupt, incremental, compound, downward, and reversal drift, with coverage changing by $-0.5$ to $+1.0$ percentage points. At matched achieved coverage, it saves $4$--$18\%$ of channel uses, while resource use in load-only and stationary cases changes by less than $0.5\%$. With $K$ samples per state and $L$ states, predictor memory is bounded at $O(KL)$ rather than growing as $O(t)$.

Partial information decomposition (PID) splits the information that two sources carry about a target into redundant, unique and synergistic parts, with nonnegative decompositions lying on a segment where redundancy measures are bounded by the positive part of the co-information and the smaller source-target information. We parameterize this segment by the overlap, the common excess of redundancy and synergy over their net balance, and show that for many of the named PID measures, their redundancy can be written as the largest positive part of the co-information over a set of reference distributions. This common form induces an ordering of those PIDs on the segment. With the smallest redundancy and synergy and the largest unique atoms of any nonnegative PID, the information-theoretic version of Hi-Fi, a feature-importance method, is the zero-overlap end of the segment for two sources, with the maximum overlap bound given by the minimum mutual information (MMI) PID. We introduce, by splitting the co-information into a pairwise maximum- entropy part and the third-order connected information, a PID that sits at zero overlap except when pairwise redundancy and triadic synergy coexist. For three or more sources the bivariate results hold on every pair of merged source groups, and among all nonnegative solutions of the lattice equations MMI and Hi-Fi still give the smallest and largest unique atoms but zero overlap on all pairs at once is often unattainable. Two inequalities across pairs are necessary in general and sufficient for three sources. For identity targets zero overlap is possible exactly when the sources' information diagram is nonnegative. In random three- and four-source distributions zero overlap on all pairs fails in 21-100% of the cases.

Music source separation extracts vocals and instruments from a mixed recording, enabling remixing, content production, and music analysis. Existing systems commonly operate on waveforms or full-band spectrograms. Waveform inputs contain many samples even in short segments, whereas a single spectrogram resolution imposes a shared time-frequency resolution trade-off across frequency ranges. We propose SubRoFormer, a Transformer-based system comprising a subband module and a residual RoFormer module. The subband module decomposes the mixture into frequency-localized, lower-rate signals for separation. It then reconstructs the target waveforms from the estimated subbands. The residual RoFormer module models temporal and cross-band relationships on temporally compressed features. It combines this context with fine-rate features through gated residual fusion to estimate source-specific complex masks. On the MUSDB18-HQ test set, training only on MUSDB18-HQ yields an average signal-to-distortion ratio (SDR) of 10.63 dB across four stems using target-specific models, achieving state-of-the-art performance among the compared systems without extra training data. With 500 additional training songs, a single model jointly separating all four stems achieves an average SDR of 11.59 dB, establishing state-of-the-art performance among the compared single-model multi-stem systems trained with extra data. Code and checkpoints are released at https://github.com/ArchiMickey/SubRoformer.

Current 4D occupancy world models use heavy cross-attention or deep recurrence to disentangle ego-motion from scene dynamics. This requires significant computational capacity. However, this effort is largely redundant since the ego-vehicle's intended future motion is provided by the downstream motion planner. We present FreeOcc, a lightweight continuous-time world model that introduces explicit $SE(2)$ geometric decoupling into a conditional flow-matching framework. By mathematically factoring out self-motion, we create a purely dynamic feature space. This space supports attention-free temporal fusion and allows physics-constrained ODEs to operate strictly on residual agent kinematics. To maximize deployment efficiency, FreeOcc introduces an uncertainty-aware blending mechanism. This routes well-observed voxels to a single-pass decoder, reserving the ODE solver exclusively for ambiguous, high-uncertainty regions. Requiring only 28.3M parameters-roughly 40% of comparable models-our optimal ensemble mode (FreeOcc-E) runs at 27 FPS. On nuScenes, FreeOcc achieves state-of-the-art forecasting performance with a 0.47 m Near Field Chamfer Distance (NFCD) and 32.1% mIoU. Furthermore, the model transfers zero-shot to SemanticKITTI. It also proves highly robust, retaining 92.8% of its ideal mIoU under significant sensor packet loss and timestamp jitter.

Rare genetic disorders are difficult to diagnose due to their diversity and overlapping clinical features. Facial morphology can provide diagnostic clues, yet most diagnostic models offer limited interpretability and are not designed to support broad screening for undiagnosed disease. We present ProtoGestalt, an interpretable prototype-based framework for facial phenotyping, rare-disease screening, and syndrome identification. The method extends prototype learning by constraining learned visual features to select facial regions, allowing predictions to be explained through localized and additive evidence. We assess ProtoGestalt's ability to distinguish syndromic from unaffected faces on multiple control datasets and class prevalence values. We further evaluate it's diagnostic performance on syndromes observed during training. The learned prototype-based representations provide significant separation between syndromic and case-matched control images and retain useful performance as the proportion of controls increases. The framework also demonstrates the ability to differentiate among syndromes. Although reasonably accurate specialized models for facial classification of rare disorders already exist, ProtoGestalt offers an interpretable alternative that links predictions to anatomically meaningful facial regions, potentially supporting clinical decision-making and the discovery of phenotype-gene associations.

LLM routing aims to reduce inference costs by assigning easy queries to less expensive models and reserving stronger models for cases in which they are needed. However, we find that existing routers often violate this principle under loose cost budgets: as the budget increases, they increasingly select the strongest and most expensive model even when less expensive feasible candidates provide comparable or identical utility. We refer to this behavior as strong-model over-selection under loose budgets. Our analysis shows that ambiguity alone does not explain this phenomenon. Substantial overpayment can arise when differently priced models offer equivalent or near-equivalent utility and the router retains a persistent model-level preference for costly models; budget relaxation exposes this preference by making premium candidates feasible. Motivated by this diagnosis, we propose EquiRouter, a decision-aligned router that learns query-dependent model rankings, preserves meaningful utility differences, and explicitly prefers lower-cost models among equivalent candidates through a Cost-Aware Ranking Objective and a lightweight Query--Model Interaction Representation. Experiments across diverse text, multimodal, continuous-utility, and large-model-pool settings show that EquiRouter consistently improves cost efficiency while maintaining competitive peak and whole-curve performance. Code is available at https://github.com/LAMDA-Model-Reuse/EquiRouter.

Predicting spatial gene expression from histology images could scale spatial transcriptomics (ST) to image-only cohorts, but conventional histology-based ST prediction is trained and evaluated mainly by per-gene spatial-profile reconstruction. This objective is misaligned with a key downstream use of ST: differentially expressed gene (DEG) discovery, where genes are ranked for a biological or morphology-defined contrast by evidence of between-group expression differences. We formulate image-based differential expression ranking (IDER), which asks whether predicted expression profiles preserve the contrast-specific ranked gene list obtained from measured profiles. IDER compares gene rankings induced by differential-expression statistics, rather than raw expression magnitudes or per-gene spatial correlations. We further introduce a differentiable IDER objective that aligns these statistics across genes and can be trained with morphology-derived proxy contrasts without predefined biological group labels. Experiments on public ST datasets show improved DEG-ranking agreement and pathway-enrichment overlap over conventional reconstruction objectives, including morphology-derived and pathologist-annotated tissue-region evaluations.

Self-supervised training is monitored without labels, through the loss and statistics of the learned features. We report a case where we know the encoders lost part of what they had learned, so we can check whether these signals noticed. We trained LeJEPA encoders on synthetic images with known generating factors. In nine of ten seeds, the readability of those factors fell abruptly and stayed down, nearly always along the same direction in factor space. Meanwhile the loss kept falling, and most of the eleven label-free signals we tracked kept improving and rated the declining runs better than runs that did not decline. Standard change-point detection rarely flagged the decline near its onset, and selecting checkpoints by these signals mostly chose ones after it.

Repeated directed reachability queries arise in online path planning, task-agent allocation, and swarm control on static grid maps, where reachability is asymmetric and full transitive closure is storage-prohibitive. We introduce H-BRICK, a Hierarchical Boundary Reachability Index with Compressed Kleene closure for exact cell-to-cell queries on directed grid graphs. H-BRICK partitions the graph into tiles, computes a local Boolean Kleene closure within each tile, and projects it onto the tile's boundary ports. The resulting directed port-to-port summaries are then composed recursively across shared interfaces up to a single root covering the graph. We prove algebraically that the construction is exact: in the Boolean Kleene algebra of binary relations, every composed summary equals the boundary projection of the regional reflexive transitive closure, and every cell-to-cell query agrees with reachability in the original graph. Experiments on directed grid graphs across diverse map topologies compare H-BRICK against directed search and component-based baselines, and compare its storage requirements against full transitive closure. A batched reachability evaluation further demonstrates that sharing endpoint lifts across many-source/many-target queries amortizes hierarchical traversal and, at k=64, yields a 5.0x-22.9x throughput improvement over the scalar query loop. The results show that hierarchical boundary compression is most advantageous in query-intensive spatial graphs where online search traverses large reachable regions and the hierarchy maintains compact boundary interfaces.

Large language models are classically trained in stages: pretraining on raw text followed by post-training for instruction following and reasoning. However, this separation creates a fundamental limitation: many desirable behaviors such as safety, factuality, overall generation quality, and reasoning ability are only added at a late stage, even though the patterns learned earlier strongly shape a model's capabilities. To tackle this issue, we introduce a new way to pretrain and mid-train models that incorporates these behaviors earlier. We utilize an existing strong, post-trained model to both rewrite data and to judge policy model rollouts, thus introducing thinking and reinforcement earlier in training. We further improve reasoning at all stages by introducing the concept of self-improving post-thinking: improving the model's reasoning and self-judging ability by reasoning after generation. In our experiments, we show these approaches can give strong gains across multiple axes.

Low Earth Orbit (LEO) satellite-IoT systems increasingly provide connectivity to mobile assets in remote areas, which makes their resilience to radio-frequency jamming a growing concern. Timely jamming detection, i.e., before complete link failure, has been studied for terrestrial networks and Global Navigation Satellite Systems (GNSS), but it remains unclear whether these approaches are effective on LEO links, where propagation varies rapidly, the received signal is weak, and jamming should be detected from the messages that are still correctly received. This paper investigates the extent to which jamming of Iridium LEO satellite-IoT links can be detected from the physical layer of correctly decoded messages. To this aim, we collect and publicly release a real-world dataset of Iridium messages under 47 jamming conditions, spanning two propagation environments, static and mobile receivers, and different jammer distances, powers, and signal types. We use this dataset to analyze three one-class autoencoder-based anomaly detectors operating on physical-layer information and compare them against two signal-quality estimators commonly used as jamming indicators in GNSS-based applications. Our results show that detectability is governed by the rise of the receiver noise floor, through which jammer power, distance, and receiver mobility act. With 1000-sample chunks, i.e., 40 ms of received signal at 25,000 samples per second (about four correctly decoded messages), the autoencoders reach an AUC of at least 0.84 whenever a noise jammer raises the noise floor by at least 9 dB, whereas the $C/N_0$ and SNR estimators fail, since received messages, although correctly decoded, retain a high SNR under jamming. Therefore, timely jamming detection for LEO satellite-IoT links like Iridium should rely on physical-layer information beyond the signal-quality metrics adopted in the GNSS domain.

Self-hosting AI is often presented as a way to reduce inference costs and increase control, but deployment introduces its own costs. We report our experience deploying a hybrid AI stack at Grayhat Developers, a small software company with an AI-intensive development workflow. Our initial motivation was to counter the growing cost of commercial AI usage: Claude/Codex incurred approximately \$325 per month, with more than 25.93 billion recorded tokens. We experimented with local open-source models as replacements for commercial coding agents, but after learning that their performance and inference speed were insufficient, we adopted a hybrid approach in which commercial agents remained responsible for complex tasks while local models handled lower-complexity tasks, organizational information retrieval, and synchronization with self-hosted tools including Outline, Plane, and ERPNext. We estimated that approximately 30--40\% of our commercial AI workload was associated with context transfer, documentation synchronization, task management, and related operational coordination. By moving this work to local models, we estimate that approximately 90\% of this coordination workload could be handled without commercial inference, corresponding to an estimated 27--36\% reduction in overall commercial-AI workload or cost-equivalent usage. Our first internal deployment nevertheless failed to achieve adoption because it introduced UX friction. Replacing it with a centrally configured deployment substantially improved feedback and adoption. We conclude that local inference is most useful not as a wholesale replacement for commercial AI, but as part of a hybrid architecture that assigns tasks according to their capability requirements, cost, context demands, and operational sensitivity.

We give a $n^{O(\ell)}$-time algorithm for solving semirandom planted instances of a $k$-ary Boolean constraint satisfaction problem (CSP) with $m \gtrsim \left(\frac{n}{\ell}\right)^{k/2} \ell \, \mathrm{polylog}(n)$ constraints, for any choice of the runtime parameter $\ell$. Up to $\mathrm{polylog}(n)$ factors, this matches the known runtime vs.\ number of constraints trade-off for refuting semirandom CSPs, as well as the trade-off for solving fully random planted CSPs. Prior work on designing algorithms for semirandom CSPs follows a high-level blueprint of decomposing the constraint hypergraph $\mathcal{H}$ of the instance into pieces, each with an explicit pseudorandom property that allows the standard algorithm for random CSPs to succeed on every piece. We depart considerably from this framework and instead implicitly \emph{learn} such a decomposition, without appealing directly to explicit properties such as expansion. More generally, we construct such a learning algorithm from the assumption that semirandom $k$-XOR has Sum-of-Squares certificates of refutation. This yields the broader implication that any Sum-of-Squares refutation algorithm for semirandom $k$-XOR gives a corresponding algorithm for solving semirandom planted CSPs, and instantiating this reduction with known Sum-of-Squares algorithms for refuting semirandom $k$-XOR gives our main algorithm. To the best of our knowledge, this is the first reduction from search to refutation for (semi)random CSPs.

State-space models (SSMs) recover low-dimensional latent dynamics from high-dimensional observations, but latents are generally not identifiable from observational data alone. Interventional state-space models (iSSMs) use perturbations to break observational symmetries and identify parameters up to permutation, scaling, and shift. However, prior work restricts identifiability to linear latent dynamics and piecewise-linear emissions, excluding many behaviors characteristic of neural circuits. We extend iSSMs to a substantially broader model class: we prove that (i) affine identifiability holds under general injective emission functions and Gaussian latent transitions, and (ii) with interventions, identifiability up to permutation, scaling, and shift is preserved under general nonlinear latent dynamics. We validate our theory on simulated FitzHugh-Nagumo dynamics, and then demonstrate two applications: adjudicating between competing models of decision-making; and predicting responses to unseen micro-stimulation in macaque and human neural recordings, with proof-of-concept generalization to a held-out stimulation site in the latter. Finally, simulations yield practical guidance for designing stimulation protocols that improve identifiability.

World-action models (WAMs) jointly model future traffic states and driving actions for end-to-end autonomous driving in intelligent transportation systems. However, their action learning remains dominated by expert-trajectory imitation, leaving unsafe alternatives close to expert trajectories insufficiently constrained. To improve WAM robustness in safety-critical scenarios, we propose NegWAM, a safety post-training framework for unified WAMs that learns explicit safety boundaries from near-expert failure trajectories. NegWAM constructs negative trajectories that are geometrically close to the expert trajectory yet carry safety risks, uses them as repulsive anchors to steer predictions away from dangerous behavior, and encodes them as scene-conditioned rider tokens to learn scene--trajectory compatibility. During training, NegWAM jointly optimizes the flow-matching and near-expert negative objectives, while the auxiliary rider branch is discarded at inference, incurring no additional overhead. On NAVSIM v2, NegWAM achieves 40.4 EPDMS on navhard, a 4.1% relative improvement over the strongest compared scorer-free WAM baseline, and achieves 93.3 EPDMS on navtest. These results demonstrate that NegWAM achieves strong performance in safety-critical scenarios while preserving routine-driving performance. Zero-shot closed-loop evaluation on HUGSIM further demonstrates cross-benchmark generalization. The code is available at https://github.com/Zizizi-hao/NegWAM.

Humanoid rigging for modern characters with complex garments remains challenging due to the need for dense auxiliary skeletons and precise skinning control. Existing automatic methods primarily recover rigs from geometry, providing limited artist control over auxiliary articulation and often lacking explicit semantic or kinematic-chain structure for dense garment bones, leading to instability and weight bleeding. We present TailorRig, a framework that reformulates garment-aware humanoid rigging as a structured, controllable full-rig generation problem. This formulation shifts the goal from automatic rig recovery to artist-driven rig authoring, where both body-template compatibility and auxiliary articulation are explicit design choices. TailorRig exposes two explicit axes of artist control within a unified framework: a user-specified body skeleton template that preserves body-rig compatibility, and artist-provided semantic labels that drive an autoregressive generator to synthesize requested category-specific auxiliary bone chains. To robustly assign skinning weights under dense skeletal configurations, we further introduce a hierarchical chain-to-bone formulation: vertices are first associated to a kinematic chain, and bone weights are then refined within that chain, reducing weight bleeding between spatially adjacent but semantically unrelated structures. We evaluate TailorRig on a diverse set of characters and introduce Human-Mix, a large-scale dataset of over 38,000 humanoid models curated for garment-aware rigging. Experiments show that TailorRig improves skeleton accuracy, deformation stability, controllability, and practical usability over existing methods. Project page: https://xxxx0601.github.io/trig/ ; code: https://github.com/Tencent/TailorRig ; dataset: https://huggingface.co/datasets/tencent/TailorRig

Field devices deployed today will still be producing records when quantum computers may break their signatures, so systems that anchor IoT data on blockchains need a post-quantum (PQ) authentication path now. We measure its cost on Ethereum under the current execution rules, including the EIP-7623 calldata floor and the EIP-7825 transaction cap, and under the calldata repricing scheduled for the Glamsterdam upgrade. We combine device-side measurements of ML-DSA, Falcon, SLH-DSA, LMS and XMSS, reproduced costs of open-source Solidity verifiers, a new standard-compliant LMS/HSS verifier for SHA-256 and SHA-256/192, and complete transactions on an Osaka-rule EVM, priced with today's verifiers and with proposed precompiles. Three results stand out. Signing every record costs 8.7 times as much as ECDSA even with an ML-DSA precompile, because the calldata floor rather than verification sets the price. When each device signs a Merkle root over 100 records, a standard LMS verifier available today costs about 3,350 gas per record, against 245 gas for ECDSA in the same design, and it is unaffected by the scheduled repricing, which raises floor-bound designs by up to 1.9 times. With the proposed precompiles, ML-DSA-44 costs 1,031 gas per record, and a hybrid ECDSA and ML-DSA signature adds only 2.5%. Post-quantum device authentication is therefore deployable on Ethereum without protocol changes, provided that the latency of batching and the data-availability properties of the chosen variant fit the application.

Many quantum software development kits provide a suite of circuit optimisation passes. These passes have been highly optimised and tested in isolation. However, the order in which they are applied is left to the user, or else defined in general-purpose default pass sequences. While general-purpose sequences miss opportunities for optimisation which are particular to individual circuits, designing pass sequences bespoke to particular circuits requires exceptional knowledge about quantum circuit design and optimisation. Here we propose and demonstrate training a reinforcement learning agent to compose optimisation-pass sequences. In particular the agent's action space consists of passes for two-qubit gate count reduction used in default PyTKET pass sequences. For the circuits in our diverse test set, the (mean, median) fraction of two-qubit gates removed by the agent is $(57.7\%, \ 56.7 \%)$, compared to $(41.8 \%, \ 50.0 \%)$ for the next best default pass sequence. In doing so our model demonstrates the ability to forgo reward to improve the final two-qubit gate count. Our approach clearly separates the problem of designing circuit optimisation passes from the problem of using them optimally in practice. We expect this separation of concerns to facilitate innovations in both areas.

High scores on public data-cleaning benchmarks can reflect knowledge of familiar entities, making it difficult to assess how well an LLM repairs unfamiliar data. We present CoRE (Cleaning or Recalling), a protocol comparing public benchmarks with fresh real records and paired name surrogates built from English words or syllables. The surrogates retain error positions and non-identity values while transferring the benchmark's error mechanisms. Across 21 controlled tables and eight LLMs, repair performance varies substantially with entity names on Hospital and, for commercial models, Movies. Renaming removes some information needed for repair, such as a film's identifier mapping. It also affects repairs whose candidates remain available: replacing entity names with word-based surrogates lowers DeepSeek-V4-Pro's year accuracy from 91% to 18%. On Hospital and Movies, commercial models score higher on fresh records than on word-based surrogates. A screened continuation probe finds no positive evidence of benchmark-row memory in the tested models. Familiar facts can conflict with the task. Even with explicit table-evidence instructions, commercial models rewrite assigned directors at rates of 0.000-0.404 on known films. Reference scoring can also disagree with external records. On Rayyan, an ISSN correction matching a title-matched PubMed record receives no credit, whereas 94% of TabClean's credited repairs reproduce the reference date transformation. We provide controlled tables, scoring code and materials for offline replay.

Generative creative interfaces routinely report the user's position within a space of possibilities. Such displays imply that the quantities shown were measured from the artefact; in a growing class of systems they are instead predicted by a language model describing its own output, and the two are indistinguishable on screen. Most creative domains lack an agreed measurement against which such a self-report can be checked; audio admits one. In a measurement study of a working LLM-driven timbre interface, what a user can cause proves separable from what the interface reports. The first is exactly characterisable: the reachable set is a zonotope in parameter space whose dimension increases by one per sound placed, so the space the model may act within is demonstrably the user's own. The second is inaccurate. The displayed position differs from measurement by approximately three times the width of that region, and the error has two independent sources: model-predicted coordinates, whose accuracy varies twofold between the two models evaluated, and the interpolation step, which arises from combining synthesiser parameters and is unaffected by model quality. End to end, 27 of 36 refinement instructions produced no perceptible change, and under the weaker model the facility performed no action while reporting convergence. Effective agency (what a user can cause) and epistemic agency (whether they can know it) are therefore separable and independently engineered: structurally enforced constraints survive model failure, whereas guarantees resting on the model's own report do not. The failure mode that remains is silence presented as success. We close with a short protocol for building and auditing such displays.

Vision-based rover autonomy requires integrating perception, localization, planning and platform-specific control under limited sensing and computing resources. We present an open-source ROS 2 stack for following directional arrows, reaching a terminal cone and returning autonomously to the start. It connects YOLOv8n detection and geometric direction, range and bearing estimation with RGB-D SLAM, wheel-inertial state estimation, kinematics-aware planning, behavior-tree mission management and STM32 actuation. Standard interfaces separate reusable autonomy from the hardware layer; a proportional path-following plugin supports low-speed rover motion. Public resources include software, configuration, a labeled dataset, trained weights and evaluation logs. On 478 held-out images, the detector achieves mAP@0.5 of 0.720 under the Ultralytics evaluator and outperforms a classical OpenCV baseline under a shared evaluation protocol. The reported perception latency is 25 ms on a Jetson Orin NX. Three missions, one indoor and two outdoor, demonstrate marker-guided navigation and autonomous return without operator motion commands or reported collisions. Field results document integration failures and corrective choices; repeated instrumented trials remain necessary to quantify physical navigation errors and reliability.

Video compression at ultra-low bitrates remains challenging due to the loss of perceptually important details under limited bit budgets. Recent generative video compression methods leverage pretrained video priors to improve perceptual quality and temporal coherence, but many follow a compress-then-generate paradigm, where generative priors are primarily exploited during reconstruction. Moreover, representations learned for generation and reconstruction are not inherently optimized for entropy coding. In this work, we propose \method, which adapts pretrained video representations into compression-oriented representations while preserving their reconstruction capability. Specifically, we adapt the pretrained causal video VAE of MiniMax-H3 into a discrete video tokenizer and develop a small autoregressive entropy model that captures local and global token dependencies, with grouped prediction for parallel decoding. To further improve coding efficiency, we jointly optimize the tokenizer and entropy model using a differentiable rate surrogate, allowing rate gradients to adapt token assignments toward lower coding costs while preserving reconstruction quality. Experiments on UVG, MCL-JCV, and HEVC Class B demonstrate superior perceptual rate--distortion performance, achieving BD-LPIPS savings of 82.2\%--87.5\% and BD-DISTS savings of 78.3\%--85.0\% relative to GLC-Video.

For perceptive humanoid locomotion, the relevance of terrain geometry depends on the robot's body state and how both evolve under its actions. Existing work has advanced terrain-aware sensory fusion and cross-modal prediction, improving how visual and proprioceptive information is combined and related. However, feature aggregation alone does not specify which cross-modal relationships should be learned, while prediction between simultaneous observations does not explicitly constrain their action-dependent evolution. We present Spatiotemporal Training with Explicit Prediction (STEP), a one-stage framework that jointly supervises cross-modal relationships and action-dependent state evolution within a shared control representation. Multiple learned queries aggregate depth and proprioceptive tokens into a compact policy state. Bidirectional masked prediction uses each visible modality to predict latent features of the omitted modality, explicitly supervising terrain-body relationships. An action-conditioned predictor maps the current policy state and executed action to the next fused state, with an exponential-moving-average target defined directly in the policy representation space. Simulation comparisons show overall improvements in terrain traversal over evaluated fusion alternatives and ablations, while depth-dropout and delay experiments demonstrate the benefits of both predictive objectives under degraded visual input. Deployed on a 23-DoF Noetix E1 humanoid with onboard sensing, STEP succeeds in 98 of 100 hardware trials across stairs, a 40\,cm platform, a 70\,cm gap, stepping stones, and backward stair traversal without rearward terrain visibility.

Deep neural networks trained with different architectures, objectives, and datasets have been reported to converge on similar visual representations. However, what remains unknown is which visual properties models actually converge on and which factors may underlie this convergence. To address this, we decompose the object similarity structure of 162 diverse vision models into a small set of non-negative dimensions. To determine universal versus model-specific dimensions, we then estimate how often each dimension reappears across models. In contrast to model-specific dimensions, universal dimensions are more interpretable and more strongly driven by conceptual image properties, indicating the relevance of interpretability and semantic content as implicit factors driving universality across models. Differences in architecture, objective function, training data, model size, and model performance do not explain the emergence of universal dimensions. However, models with more universal dimensions also better predict macaque IT activity and human similarity judgments, suggesting that universality reflects representations relevant to biological vision. These findings have important implications for understanding the emergent representations underlying deep neural network models and their alignment with biological vision.

Tree-species composition shapes forest productivity and how forests respond to climate disturbances such as drought, windthrow and insect outbreaks. Yet most national forest maps assign each pixel a single dominant species and cannot show the share of accompanying species. We mapped the basal-area shares of ten main tree species and a residual Other class across Switzerland at 30 m resolution. A deep-learning model combining 0.1 m aerial RGB imagery, airborne laser scanning (ALS), annual AlphaEarth satellite embeddings, and climate variables was trained on basal-area observations from 3,992 Swiss National Forest Inventory (NFI) sites. In spatial cross-validation, the model reached a macro-$R^2$ of 0.485 across the 11 classes. On an independent forest-reserve dataset, the map reached a macro-$R^2$ of 0.552 for individual 30 m pixels and 0.733 for plots, and identified the dominant class in 73.2% of pixels and 81.1% of plots. The map shows maple and ash as widespread minor components that dominant-class maps omit. It also reproduces the elevational sequence from broadleaved lowland stands through spruce to larch and Arolla pine near the treeline. By resolving the species mixture within each 30 m pixel, the map complements field inventory statistics with spatially continuous information that can support forest monitoring, the management of mixed stands, and climate-adaptation planning across Switzerland.

As large neural models have become better at language tasks, researchers are increasingly building multi- and omnimodal models that handle more than one modality of data. One example is the expansion of speech recognition models to audio-visual data, the goal being noise mitigation and multimodal subtitling. While performance and bias have been studied extensively in the single-modality regime, it is unknown how new modalities affect this, even though they produce biases in humans. We therefore propose the first bias evaluation of multimodal speech recognition, where we create videos pairing different faces with the same audio using lip-sync, and measure changes in speech transcription accuracy. We find large quality-of-service differences across mWhisper-Flamingo and Gemini models, with drops of up to 4.05 word error rate points, across self-declared gender, ethnicity, and their intersection. Our findings point to a priority for developers to evaluate, fix, and communicate such limitations, as providing more signals through additional modalities is not necessarily better, and may even lead to biased outcomes.

Existing generative models for earth observation (EO) predominantly rely on fine-tuning natural image priors, which limits their scalability and introduces perspective biases that conflict with geospatial constraints. To address this, we introduce GeoCore-9B, a 9-billion-parameter generative foundation model, which is the first of its scale to be trained from scratch exclusively on EO data. Unlike previous EO foundation models, GeoCore-9B is built upon a Flow Matching-based Diffusion Transformer (DiT) and natively conditions generation on text descriptions and continuous geospatial metadata, including ground sample distances, latitudes, and longitudes. To overcome the convergence and spatial disorientation challenges of training at this scale, we propose a Geospatial Semantic Alignment loss. This objective distills structural Earth surface priors (e.g., terrain and urban areas) from a frozen specialist teacher network, constraining the diffusion latent trajectory during training without adding inference overhead. Pre-trained on the global-scale Git-10M dataset, GeoCore-9B demonstrates strong downstream versatility. Beyond standard proxy generative tasks, we show that GeoCore-9B can be effectively adapted for practical EO applications, including highly challenging tasks such as cloud removal and SAR-to-optical cross-modal translation. Extensive evaluations confirm that GeoCore-9B establishes new state-of-the-art performance in both visual fidelity and geographic structural accuracy.