Papers

Planning en-route charging and battery-swapping infrastructure for long-haul battery-electric trucks (BETs) requires models that reflect how trucks actually operate. This paper develops a mixed-integer programming framework that jointly sites charging or swapping stations and schedules each truck's charging, swapping and mandatory driver rest, so that charging time overlaps with regulated rest instead of being added to it. Energy use is derived segment by segment from road terrain with a tractive-force model, and the truck battery is modelled as a set of independently swappable packs. Staged investment under uncertain BET adoption is formulated as a multistage stochastic program with Markovian demand and solved by stochastic dual dynamic integer programming (SDDiP) with Lagrangian cuts; we show that the Lagrangian multipliers can be bounded by each station's annualised cost without weakening the cuts. Applied to twelve freight corridors on Australia's East Coast, the algorithm jointly optimises charging and battery-swapping events as well as mandatory break events. A +/- 30% change in adoption alters the final charge-only network by only -12% to +13% of stations, with over 95% of stations built by the second stage. Hedging against uncertainty mainly changes which sites are chosen (71% overlap with a deterministic rolling-horizon model), not when they are built.

Computing-in-Memory using DRAM (CIMD) has demonstrated substantial energy and throughput gains for memory-bound workloads consisting of bulk-bitwise operations, by performing computation directly within DRAM subarrays. Realizing CIMD, however, requires a redesign of the memory controller and careful mapping of operands onto the memory arrays. Presently, accurate FPGA-based testbeds exist, but they are costly and labor-intensive, while open-source simulators are largely trace-based and cannot execute full application runs. The only full-application CIMD simulator available, included with MIMDRAM, is built on an outdated version of gem5 and its toolchains. We present gem5-CIMD, a full-application CIMD simulation framework built on the latest version of gem5. We extend the simulated instruction set from four bitwise operations to sixteen CIM instructions spanning arithmetic, relational, conditional, and utility operations across a range of data types and bitwidths (4 to 64 bits). A complete memory-management stack comprising multi-level huge-page support, a CIM-aware allocator, and a subarray-aware address-mapping scheme ensures that CIM operands are co-located in the same DRAM subarray. In addition, gem5-CIMD is aware of the DRAM refresh interval and faithfully models and schedules the mandatory refresh operations when needed. We provide a CIM standard library intended as a compiler target and evaluate gem5-CIMD on multiple case studies including an end-to-end KNN workload. The simulator sources, including the evaluated workloads, are publicly available.

A client renting a cloud quantum processor cannot easily tell whether its bitstrings came from a quantum device. Random circuit sampling certifies present hardware, but scoring each sample costs the client time $O(2^n)$. We give a protocol in which scoring a round costs $O(n)$. It uses degenerate concordant ensembles: classical distributions over a product basis in which several basis states share one probability. Every concordance-preserving gate on such a state factorizes into local basis changes, a relabeling, and a block factor that rotates among equiprobable labels and commutes with the state, so it cancels from the ensemble while entangling each pure component. We call these degenerate concordant circuits (DCCs). The client holds a secret parity whose cosets are the degenerate sectors of every layer, so it builds the circuit and predicts the ensemble with one inner product per layer; the secret never leaves the client. Inputs are uniformly random; the client privately weights each round by its input's eigenvalue in the hidden state. An input-output agreement test and a weighted collision test compare the server's samples with the non-flat output distribution. We prove completeness, client efficiency, and soundness in an ideal model where the compiled circuit reveals nothing about the secret, and show that a classical server holding it passes. For the explicit construction we prove that the single-qubit symmetry test of efficient concordant simulations fails on non-product inputs with unbiased marginals, and find that a published local-basis finder halts on every layer tested, each run saturates the maximal Schmidt rank, the circuit is not Clifford, and matrix-product spoofers are rejected. Other blocks and permutations are possible. As these tests are not exhaustive, soundness is conjectured for constructions whose public data carry no signature of the secret.

We determine the first extension groups between simple restricted modules for the restricted simple Lie algebras of Cartan type over an algebraically closed field of characteristic $p\ge5$, which answers the restricted case of a problem of Benkart and Feldvoss. Extensions from parabolically induced modules are computed by finite restricted-cohomology complexes of the grading parabolic, and for the exceptional modules we obtain closed formulas in all four families. Neither the trivial module nor any nontrivial exceptional simple has self-extensions. In characteristic seven we determine the quiver of $S(3;1)^{(1)}$, which has 49 vertices and 566 arrows.

Multi-task manipulation policies differ in architecture, scale, and pretrained priors all at once, so published comparisons cannot attribute performance to any single component. We argue that most of it comes from the visual representation, and that parameter scale and generative priors are largely incidental. To test this, we build CoRP (Compressed Representation Policy), a deliberately compact policy (48.9M parameters, no vision-language model and no video-generative prior) that factorizes into a representation extractor and a flow-matching action generator. It reaches 97.0% on LIBERO and 75.78%/73.36% on RoboTwin 2.0 Clean/Randomized, matching systems 40.9-163.6x larger. Holding the action generator fixed, we then vary one extractor property at a time. Pretrained initialization is decisive: a random ViT-S/14 drops to 78.1% and an ImageNet ResNet-34 to 74.5% on LIBERO. Pretraining alone is not enough, as freezing the encoder costs 19.8 points. Compression matters as much: resampling each view to 48 tokens beats passing all patch tokens (97.0% vs 83.2%), and a variational information bottleneck over those tokens is worse than a hard token budget, cutting LIBERO-Goal from 95.8% to 33.0% by suppressing the instruction-dependent token selection the policy relies on. Language conditioning contributes only where the observation leaves the goal ambiguous (LIBERO-Goal: 9.2% to 95.8%), while on RoboTwin 2.0, where observations are unambiguous, removing it slightly improves success. Therefore, we argue that a compact policy works when its representation is pretrained, task-adapted, and compressed. Project page: https://corp-policy.github.io/

Recent text-guided audio FX research has focused on translating natural-language descriptions into parameters or chain configurations within existing FX systems. However, comparatively little attention has been paid to how the internal control space of an individual FX processor can itself be constructed. To address this gap, we propose CTAG-FX, which functionally reinterprets the roles of 78 parameters in a text-conditioned synthesizer configuration as controls of a tone-shaping FX processor. Each synthesizer configuration defines a fixed tone-shaping processor that can be repeatedly applied to new audio inputs rather than producing only a one-off rendered output. The text-conditioned synthesizer configurations are generated using A&R-CTAG, a retrieval-enhanced extension of CTAG. We evaluate CTAG-FX through signal-level analysis, together with a scrambled-mapping ablation in which parameter-to-control assignments are randomly permuted to assess the contribution of the proposed role assignment. Role-based mapping produces prompt-distinct spectral and nonlinear behavior, whereas scrambling reduces this prompt-specific differentiation and yields a more prompt-insensitive nonlinear profile. Overall, these results suggest that synthesizer parameter spaces can serve not only as sound-generation spaces but also as design resources for constructing new audio-FX control spaces. Audio samples are available at https://taylor3527eg-hub.github.io/ctag-fx-platform-demo/.

We propose a new variant of a biased random walk on a Galton-Watson tree, in which some vertices have infinitely many children. These vertices induce a natural regeneration structure, which allows to study differentiability and also monotonicity of the speed of the random walk in the bias parameter. Additionally, we show that the speed in this model is the limit of the speed of random walks on appropriate locally finite trees.

Zonal-flow organization can modify the relation between heat flux and temperature gradient, but how fast ions affect this organization and the accompanying transport response remains less well characterized. We use nonlinear, global gyrokinetic simulations based on KSTAR fast-ion-regulated enhancement (FIRE) mode to compare cases with and without fast ions during profile relaxation across five initial main-ion temperature gradients. We observe that fast ions induce larger zonal-flow and shearing-rate amplitudes, together with persistent shear layers. At the reference profile based on the experiment, these shear layers coexist with temperature-gradient corrugations qualitatively consistent with $E\times B$ staircases. When fast ions are present, inward and outward ion heat fluxes coexist at different radii, which causes the average heat flux to be smaller than when fast ions are absent. This redistribution of ion heat transport may be relevant to understanding the improved ion confinement observed in FIRE mode. In particular, we note a transition-like response in the presence of fast ions, whereby the heat flux decreases as the ion-temperature gradient increases, in contrast to the conventional Fick's law trend without fast ions. This behavior that deviates from Fick's law suggests that FIRE mode may represent a bifurcation to a distinct transport regime rather than a continuous high-performance extension of L-mode. These findings highlight the importance of the initial temperature gradient and global zonal-flow organization when assessing fast-ion effects on ion heat transport.

Text-driven 3D Gaussian editing commonly does not distinguish the editing reliability of rendered views, although different viewpoints provide supervision of substantially different quality. Views that clearly show the scene and match the edit instruction provide reliable guidance, while less informative views may weaken the edit when all views are treated equally. We present View Matters, a view-importance-aware framework that conducts editing around reliable keyframes. Keyframe Importance Estimation (KIE) identifies reliable views using geometric visibility, semantic distinctiveness, and edit relevance. Keyframe-Guided Editing (KGE) then propagates their editing signals asymmetrically to non-keyframes without noisy reverse influence, while Importance-Aware Optimization (IAO) preserves this reliability preference during 3DGS optimization. Across 23 scene-prompt pairs, View Matters achieves the highest average CLIP text-image similarity of 0.2822 and directional similarity of 0.2564 among the evaluated methods, with a four-minute editing time. Additional adjacent-view analysis indicates that the fidelity-oriented editing process maintains cross-view coherence.

Existing latent-based watermarking methods for diffusion models have overestimated their robustness to image distortions, including geometric transformations such as rotation, scaling, and translation (RST). Moreover, this paradigm of watermarking approaches may suffer from inherent limitations arising from the domain in which the watermark is embedded. In this paper, we provide the first theoretical analysis explaining why these methods lack invariance to perturbations. By relaxing the invariant relation, we derive a maximum perturbation bound that characterizes the relationship between pixel-space perturbations and their corresponding effects in latent space. In addition, we present the first analytical formulation that captures all components of practical detection mechanisms. Finally, we conduct experiments to validate the theoretical findings and the limitations of latent-based watermarking methods. Our theoretical and empirical results indicate that, under the current design paradigm, latent-based watermarking methods intrinsically exhibit limited robustness. We conclude by providing the analytical tool and design guidelines that future research could follow.

Speed-only aerodynamic power models for multirotor propulsion cannot represent acceleration-dependent effects. This work develops a nested sequence of propulsion-power models that starts from aerodynamic power dissipation and progressively introduces a reversible kinetic-energy rate, torque-dependent electromechanical dissipation, and lumped speed-proportional dissipation. The models are identified using one subset of experiments and validated using the other on a motor-drive-propeller unit. Independent estimates of rotational inertia and aerodynamic drag complement predictive validation by assessing whether the models correctly attribute the measured power to reversible kinetic-energy exchange and irreversible dissipation and, within the latter, to aerodynamic and electromechanical losses. The results show that the reversible kinetic-energy rate is necessary but insufficient for accurate dynamic power prediction. Dissipation proportional to the squared motor torque provides the main additional improvement, while speed-proportional dissipation further prevents irreversible losses from being attributed to reversible kinetic-energy exchange. The resulting methodology provides a reusable and experimentally verifiable basis for developing and selecting dynamic propulsion-power models for multirotor systems.

Decentralized learning is highly sensitive to communication topology under non-IID data. Adaptive peer-selection methods can exploit local model information, but broader peer discovery may require increasingly large control state, whereas direct spectral optimization typically relies on graph-wide information. We study the intermediate setting of bounded local topology search and propose Local-First Heuristic Evolution (LFHE), a representation-driven rewiring framework whose candidate discovery and scoring use only ego-neighborhood and friend-of-a-friend (FoF) information. The structural score admits an exact interpretation through graph Dirichlet energy: its sum across clients equals twice the representation Dirichlet energy, which under standard linear consensus dynamics governs the instantaneous dissipation of representation disagreement. LFHE combines this state-dependent structural signal with early exploration and degree control, while algebraic connectivity remains an offline graph diagnostic. Under bounded sparse degree, its FoF candidate state remains local rather than expanding toward population-wide peer tracking. Across four image, speech, and text benchmarks, LFHE achieves competitive decentralized learning performance. Matched-protocol controls identify the structural term as the principal empirical topology-selection signal, while comparison with broader peer discovery exposes a trade-off between predictive performance and discovery-state locality. Together, these results motivate state-aware bounded local topology search between pairwise peer selection and globally informed topology optimization.

We investigate why changing the heavy-neutrino mass hierarchy can reverse the baryon asymmetry in nonthermal leptogenesis after hilltop inflation. Along the mass trajectory studied here, lowering the mass of the lighter state, \(N_1\), reverses the final asymmetry while leaving the dominant injection from \(N_2\) decays nearly unchanged. The mass change primarily delays \(N_1\) washout relative to electron–muon flavor decoherence; at fixed source and background, translating the washout history while preserving its shape and integrated strength reproduces more than 99% of the yield changes from the nonthermal \(N_2\) source at two physical target masses. Decomposing the \(N_2\) CP source shows that the projector-aligned component remains positive under both washout histories, whereas the initially trace-free component reverses its propagated contribution and accounts for about 99.7% of the timing contrast in the fixed-source comparison. Rapid damping of \(e\mu\) flavor coherence largely removes this contrast, while rapid chemical equilibration does not. Later washout therefore acts on a more decohered flavor state, reducing the positive \(e\mu\) contribution and allowing negative diagonal contributions to determine the final sign. At a mass recomputed to reproduce the observed baryon yield, a 10% increase of the muon collision operator reverses the sign in a fixed-input sensitivity test.

Continuous authentication (CA) strengthens session security by repeatedly verifying the user during device interaction, yet it inherently relies on highly sensitive behavioral traces (e.g., fine-grained touch dynamics) that are often outsourced to cloud/edge services for scalable inference. This raises a fundamental privacy-in-use challenge: protecting behavioral features during computation, not only in transit or at rest. Fully homomorphic encryption (FHE) offers a principled solution, but deploying modern CA models under FHE remains difficult due to non-linearities and attention-style operations that incur high ciphertext cost. We propose FBAN, a TFHE-compatible Bottleneck Attention Network and an end-to-end encrypted CA framework. FBAN is designed for integer-only execution via a two-stage pipeline (floating-point pretraining followed by quantization-aware training) and is compiled into TFHE circuits for homomorphic inference. We further specify a client-server protocol with session-bound blinding and decrypt-and-return verification, enabling the server to authenticate users without observing raw behavioral features. We provide a cryptographic security analysis against an honest-but-curious server under TFHE IND-CPA security, and formalize resistance to replay and impersonation without the TFHE secret key. Experiments on two public touchscreen datasets demonstrate that FBAN achieves strong authentication utility under encrypted inference while maintaining a lightweight model footprint, with TFHE parameters instantiated at $\geq 128$-bit security.

Automated Program Repair (APR) with language models is usually evaluated by whether a generated patch passes the test suite, which can hide differences in maintainability, security, and computational cost. We propose a Weighted Quality Index (QI), inspired by the ISO/IEC 25010 software quality model, that combines functional correctness, maintainability, security, and generation efficiency under configurable weighting schemes. We evaluate three dense Qwen2.5-Coder models (3B, 7B, 14B) and the 16B-parameter DeepSeek-Coder-V2-Lite Mixture-of-Experts (MoE) model (2.4B active parameters) on 40 QuixBugs and 90 Defects4J bugs, all run locally on identical hardware to control for infrastructure effects. Model rankings change with the weighting scheme, showing that single-metric evaluation can hide trade-offs. The MoE model shows almost no statistically significant difference in correctness from the 7B and 14B dense models (McNemar's exact test) while using 3-6 times fewer active parameters, whereas correctness increases significantly across the three dense scales. These results suggest that active parameter count can be a more informative lens than total parameter count for sparse code models.

Caloric solid-state refrigeration offers an energy-efficient alternative to vapor-compression cooling, but many of the most promising materials, including the magneto- and elastocaloric Heusler alloys, are intrinsically brittle and fail under the cyclic loading required in operation. Here, we introduce mechanical confinement via powder-in-tube (PIT) processing as a general strategy to overcome this limitation. By embedding a granular Heusler alloy core of 87 % packing density within a ductile steel sheath (2.8 mm outer diameter, 0.3 mm wall thickness), the composite confines the brittle powder and enables effective load transfer between particles. While the core remains inherently brittle, the surrounding sheath prevents catastrophic fragmentation and turns brittleness from a disqualifying property into a manageable one. We demonstrate that this approach enhances mechanical durability by several orders of magnitude, with the composites sustaining 100,000 load cycles at 250 MPa without structural failure and tolerating overload stresses up to 700 MPa. Confinement also alters the transformation itself, and to resolve this behavior in these mechanically complex systems, we developed a simultaneous in-situ measurement technique combining strain and AC magnetic susceptibility under load. It reveals that the internal stress field in the core is strongly heterogeneous. As a consequence, the macroscopic strain decouples from the transforming phase fraction, and the martensitic transition cannot be tracked by mechanical data alone. Mechanical confinement thus emerges as a design principle for fatigue-resistant caloric materials, and the combined in-situ method as a characterization tool for composite, porous, and other mechanically heterogeneous caloric systems.

The Worley-Sagan insertion and Haiman's mixed insertion are insertion algorithms for shifted Young tableaux, and each of them gives a Robinson-Schensted-type correspondence between the symmetric group of degree $n$ and a set consisting of certain pairs of same-shape shifted Young tableaux with $n$ cells. It is a known fact that these two insertions are dual to each other. Our purpose is to give an extension of these two insertions without losing the duality relationship. The extended ones will be insertions producing pairs of shifted tableaux from colored permutations. Our extension of the Worley-Sagan insertion is different from the restriction of Sagan's own "Knuth version" to colored permutations. In proving the duality between our extended insertions, we "embed" them into Shimozono and White's doubly mixed insertion for unshifted tableaux by "doubling" shifted tableaux and use the self-duality of the doubly mixed insertion shown by Shimozono and White.

Agentic VLM pipelines increasingly pass the same static visual evidence through multiple specialist agents and tools. This design creates an orchestration-level redundancy mode: semantically unchanged images are repeatedly reconstructed as image-conditioned requests at the VLM API boundary. We call this phenomenon visual orchestration tax and develop a measurement-to-certification framework for visual evidence reuse in agentic VLM pipelines. The audit side defines $\mathrm{M1}_{\mathrm{trace}}$ to count raw visual-evidence touches and M2 to measure structural touch redundancy, with query-level distributions, bootstrap confidence intervals, and paired quality tests. Across SeeingEye and MAMMQA on chart, document, general-VQA, and multi-modal-QA tasks, audits reveal 66.8-75.6% visual-evidence touch redundancy, and every audited query exceeds the predefined gate. The certification side introduces SharedVisCache, a contract-aware evidence reuse hook keyed by image content, preprocessing fingerprint, and encoder assumptions. On SeeingEye, contract validation certifies 75.0-75.5% repeated touches as reusable while preserving 350/350 output strings and $\Delta\mathrm{M5}{=}0$. At the physical layer, certified hits reduce $F_{\mathrm{vision}}$ from 800 to 200 in ChartQA-200 trace replay and from 200 to 50 inside live SeeingEye translator-stage physical integration, preserving 800/800 replay strings and 200/200 integrated call outputs. The results position visual reuse as a measurable, behavior-preserving property of agent orchestration and define an agent-layer contract that makes backend prefix or token reuse semantically interpretable.

The growing importance of intraday trading in Europe, driven by the increasing penetration of renewable energy sources, has led to higher volatility and periods of market stress. Understanding how order flow behaves under varying liquidity conditions requires models that adapt to the state of the market. This paper proposes a multivariate state-dependent Hawkes process to model the flow of orders across different liquidity regimes in the intraday continuous electricity market. The observed market states are identified using a proposed Liquidity Stress Index (LSI), which quantifies stress using the bid-ask spread in the limit order book, order book volume, and mid-price volatility. Transitions between states induce changes in the intensity parameters, resulting in piecewise dynamics that reflect regime-specific order-flow behavior. The model is calibrated on January-March 2024 Dutch intraday continuous cross-border intraday (XBID) hourly order data from EPEX SPOT. The estimated dynamics show that order flow is strongly self-exciting and near-critical, that executed trades trigger new same-side orders while the reverse effect is negligible, and that liquidity stress mainly strengthens transaction self-excitation rather than reshaping the cross-side structure. Robustness checks over the LSI composition and the state partition show that the results are driven by the regime construction itself: alternative index compositions fit within a narrow band, three states fit better than two, and removing the states clearly degrades the fit relative to a single-regime Hawkes benchmark. The results highlight both the value of conditioning order-flow models on liquidity stress and the limits of a single-exponential specification when event frequencies are highly imbalanced.

Kuperberg conjectured in 1983 that every convex body $K$ in the plane is contained in a quadrilateral of area at most $\frac{3}{\sqrt5}\,|K|$, the extremal bodies being the affine-regular pentagons. On the basis of numerical experiments we propose a stronger conjecture of a purely polygonal nature: every convex polygon $P$ is contained in a quadrilateral of area at most $\frac{3}{\sqrt5}\,|P|$ whose sides are parallel to sides or diagonals of $P$. This chord conjecture implies the inequality conjectured by Kuperberg. For pentagons it is a theorem of Hong, Ismailescu, Kwak and Park, of which we give a short proof by area identities. We also show that the classical bound $\sqrt2$ remains valid for quadrilaterals with sides parallel to chords; the proof is similar to Ismailescu's proof of this bound, but it starts from an inscribed quadrilateral of maximal area instead of a circumscribed quadrilateral of minimal area. We close with a discussion of the difficulties in reaching the constant $\frac{3}{\sqrt5}$.

We study weighted coefficient estimates for the Laguerre transform in terms of moduli of smoothness generated by the Laguerre translation. For every integer $r\ge1$ and $1\le p\le2$, a weighted $\ell^{p'}$ norm of $\min\{1,nt\}^{r}\widehat f_\alpha(n)$ is bounded by the modulus of order $r$; for $2\le p\le\infty$ the reverse direction holds on the same weighted sequence scale. At $p=2$ these estimates yield a two-sided equivalence, and testing individual Laguerre polynomials proves optimality of the multiplier weight up to constants. The proof combines Hausdorff–Young inequalities with an averaged lower bound for the translation multiplier. As consequences we recover Jackson- and Bernstein-type estimates and the known $L^2$ approximation characterisation of the Lipschitz classes for $0<\beta<r$. We also give the corresponding one-sided coefficient-tail criteria for general $p$, together with saturation and endpoint statements.

Adaptively Implicit-Explicit (AdImEx) time stepping provides stability for large time steps for mass-conservative transport, making it attractive for the numerical representation of advection in weather and climate prediction. Existing AdImEx transport schemes become first-order accurate in the large-time-step, implicit regime and may require multiple sparse matrix solutions per time step. This work introduces AdHImEx, a new AdImEx scheme that addresses both limitations simultaneously. AdHImEx provides second-order accuracy for large Courant numbers while requiring only a single matrix solution per time step. It is a Runge-Kutta scheme that blends third-order accurate explicit time stepping and Crank-Nicolson implicit time stepping to provide numerically verified stability for Courant numbers up to 100. AdHImEx is designed for flows that are predominantly explicit, activating implicit time stepping only where large Courant numbers occur. It retains the full efficiency and third-order accuracy of the explicit time stepping in regions with small Courant numbers. In this work, it is combined with a fifth-order accurate finite-volume discretisation in space for mass conservation. A second contribution of this work is a novel stage-dependent treatment of the divergence operator that preserves constancy despite different implicit and explicit Runge-Kutta stage time steps and spatially varying implicitness, removing spurious divergence. This innovation is applicable to other AdImEx time stepping schemes. Convergence, transport, and efficiency tests demonstrate substantial improvements over previous first-order AdImEx schemes, little phase and amplitude error, accurate transport on locally refined meshes with large Courant numbers, and promising efficiency characteristics for atmospheric transport.

The mechanical properties of a metallic alloy are set by its microstructure and texture: the size and shape of its grains and the orientation of their crystals. That structure is in turn set by a recipe, the alloy composition together with the processing parameters. Alloy development runs this chain forwards, tuning the structure until a target property is met. Running it backwards, from an optimized structure to the recipe that would produce it, still relies on expert knowledge. We ask whether this backwards step can be learned. On an in-house dataset of 107 magnesium alloy extrusion conditions across 14 alloys, each with optical micrographs and an X-ray texture measurement, we compare three descriptors of microstructure and texture: conventional grain and texture statistics, a vision embedding from a pretrained image encoder, and a graph neural network on the grain network. Each is paired with prediction heads for two tasks: the alloy composition given the process (Task A), and the process parameters given the composition (Task B). Under 5-fold cross-validation, the conventional descriptors identify the correct alloy for 65% of held-out conditions, against 17% for always guessing the most common alloy, while the learned embeddings stay below 30%. The process parameters are recoverable but noisier: compared with using the composition alone, the microstructure roughly halves the temperature error. Because only a few alloys were cast and only a few press settings were used, both answers are discrete, and heads that pick from these known options, while respecting their order, worked better than heads that predict a free value.

Low-rank quantization error compensation (LQEC) recovers the accuracy lost under aggressive weight quantization by attaching a closed-form rank-$r$ adapter beside each frozen quantized weight, without any training. We show that existing compensators are limited by two shared simplifications. They calibrate symmetrically, evaluating the full-precision and compensated weights on the same activation, which yields a compensation target that is inherently high-rank – so a fixed rank budget captures only a small fraction of it. And they minimize only the second-order term of the loss, although the compensated model is not stationary: a first-order descent direction larger than the applied compensation itself remains in every layer, and no reconstruction objective can absorb it. We propose a two-stage closed-form framework that removes both simplifications. Stage 1 aligns each layer's output with the full-precision model under a Fisher-weighted asymmetric objective, concentrating the rank budget on a rank-compressible target. Stage 2 re-measures statistics on the compensated model and applies a rank-constrained natural-gradient step that absorbs the remaining first-order signal. Every adapter is the result of a single truncated SVD; backward passes serve only to collect statistics. At 2 bits under QuIP#, our method reduces WikiText-2 perplexity from 12.43 to 10.26 on Qwen3-8B and from 21.11 to 13.22 on Qwen3-4B. On the held-out C4 corpus, it recovers 51% and 84% of the gap to FP16, versus 31% and 63% for the strongest baseline, with consistent gains in the seven-task zero-shot average, at higher bit-widths, and under a distinct quantizer.

In this paper, we prove a uniform approximation theorem with interpolation for complete non-orientable minimal surfaces of finite total curvature in the Euclidean space $\mathbb{R}^n$ $(n\ge 3)$. In particular, we establish the existence of such surfaces with the complex structure of any open conformal non-orientable surface of finite topology and parabolic type.

EmoNet-Face-HQ is a benchmark of 2,500 generated portraits, each rated by four psychology experts on 40 emotion categories. It concludes that vision-language models (VLMs) on average offer no dependable improvement over random guessing and that a fine-tuned model is required. The benchmark reads each model's answer from generated text, and its zero-shot prompt demands a fixed number of emotions per face. Its own results already show that this protocol can decide the outcome: one VLM moves from chance to close to expert agreement when only the prompt changes. We keep the images, expert ratings and metric and change only how the answer is asked for and read: one yes/no question per category, scored by the probability of "yes". Scored from parsed text on the five categories experts agree on most, eleven open-weight VLMs average a quadratic-weighted $\kappa$ of 0.425 against individual experts, below the 0.468 at which two experts agree. Scored from yes-probabilities, the same models average 0.553, and all eleven agree with an individual expert better than a second expert does. The VLMs' mean is level with the benchmark's fine-tuned model (0.555), and three of them significantly exceed it without any training. Only one exceeds a stricter reference, the mean of three experts scored against the fourth (0.563). Verdicts that VLMs fail at graded facial affect ratings may therefore describe the readout more than the model. A fine-tuned model should have to beat a pretrained VLM read this way before it is called necessary, and the benchmark's own does not clearly do so.

Ambient documentation systems are rapidly gaining adoption, yet their impact on clinical note quality remains poorly characterized. We introduce MedConv, a multilingual dataset of 300 clinical encounters in English, Danish, and German, and use it alongside the Ambient Clinical Intelligence benchmark (ACI-BENCH) to compare Corti, a clinical AI platform, with two leading, accessible ambient scribe software applications built on general-purpose AI. We present a controlled clinical evaluation framework that combines entailment metrics with LLM-judged pairwise comparisons across eight dimensions adopted from PDSQI-9. Results show that Corti's API-based text-generation infrastructure is on par with or outperforms leading commercial scribes. We further show that Corti's configurable API provides the flexibility necessary to fine-tune quality dimensions for specific documentation use cases. We present the evaluation methodology and release a dataset to support future reproducible comparison of ambient documentation systems.

Migration and selection jointly shape evolutionary outcomes in spatially structured populations. Migration can vary over time as ecological shifts, such as changes in sea level, alter habitat connectivity. Here, we study the impact of periodic migration on selection within a geographically subdivided population. Using a two-deme Moran model with spatially heterogeneous selection, we analyze a scenario in which a mutant genotype is advantageous in one habitat and disadvantageous in the other. Keeping the time-averaged migration rate fixed, we vary the period, relative phase, and directional asymmetry of migration and examine the resulting steady-state mutant frequency. We find that migration periodicity generally drives the population toward neutrality: relative to constant migration, it decreases the mutant frequency when mutants are advantageous on average and increases it when they are deleterious, without reversing the direction of selection in the parameter regimes examined. This effect is strongest for out-of-phase migration, which models seasonal migration, and is further enhanced by directional asymmetry. The response to the migration period is generally non-monotonic, with an intermediate characteristic period producing the largest deviation from constant-migration behavior under in-phase migration. These results arise from the coupling between migration and local selection in Moran process: temporal modulation of migration changes not only the exchange of individuals between habitats but also the local population composition on which selection acts. Our results show that populations with identical mean migration rates can show different evolutionary outcomes depending on the temporal organization of gene flow.

COMET reports translation quality as a single number, and that number is routinely compared across target languages written in different scripts. Such a comparison assumes Script Invariance: the score should not depend on the writing system that carries the target. We test it on IndicMT Eval by re-encoding the target into Latin script, which changes orthographic form while holding content and human ratings fixed. Script identity then accounts for 22.9% of native-script COMET variance, and agreement with annotators falls in all five languages studied. We trace the effect to the tokeniser and measure it with three label-free diagnostics. The bias is two faults, not one. Scores from different scripts occupy incompatible ranges, and within a single script the metric orders translations less accurately. No order-preserving transform of the score can repair the second fault. The first is removed exactly by COMET-QN, which maps the score distribution of each (language, script) pair onto a shared reference. Pooled agreement with annotators rises from 0.300 to 0.399, which is what makes scores from different scripts safe to place on one axis, and every within-language ordering is provably preserved. A regressor over parity features recovers a further 17.1% of the lost sensitivity. The remainder belongs to the encoder, and no post-processing can reach it. We therefore recommend publishing the normalised score, the three diagnostics, and the identity of the tokeniser they were computed against, so that a reader can tell how much of a score reflects translation quality and how much reflects the writing system.

CH3OH masers that are collisionally pumped, referred to as `class I', commonly trace shock interfaces in star-forming regions, but their incidence in low-mass protostars remains poorly quantified. We use data from the COMPASS ALMA Large Program to establish the occurrence, spatial origin, and excitation of a newly detected class I CH3OH maser at 302.9 GHz towards low-mass protostars and to test whether its pumping environments exhibit signatures of shock-driven complex chemistry. COMPASS provides unbiased line surveys of 11 sources between 279.0 and 311.7 GHz at typical angular resolutions of 0.3"-0.5". Rotation diagram analysis of departures from LTE in the 302.9 GHz CH3OH excitation robustly reveals masing in seven of 11 sources, with peak brightness temperatures from a few to ~10^3 K. The brightest detection is towards HOPS 108 and is likely associated with the powerful outflow from the nearby intermediate-mass protostar HOPS 370. Excluding this case, the maser is detected in six of ten low-mass protostars (60%). All six are Class 0 sources interacting with dense ambient material, whereas the non-detections comprise three Class I sources and one isolated Class 0 core, indicating a preference for early evolutionary stages and gas-rich environments for outflow-ambient medium interactions. The maser emission is spatially coincident with thermal CH3OH along outflow cavity walls and bow shocks, but is more compact and knot-like. Among the surveyed COMs, only CH3CHO closely traces the shocked morphology of the 302.9 GHz maser, and only around HOPS 108. These results show that class I CH3OH masers in low-mass protostellar systems are more prevalent than previously recognised and preferentially occur in younger, more embedded sources. More sensitive CH3CHO observations are needed to test whether a similar relationship with shock-driven chemistry also applies in the low-mass regime.