Papers

Unconventional compensated magnets combine zero net magnetization with momentum-dependent spin polarization, but identifying them usually requires knowledge of their magnetic order. Here we show that crystal symmetry can constrain unconventional magnetic character before the magnetic ground state is known. Starting from a non-magnetic crystal structure and a specified magnetic sublattice, we generate symmetry-compatible compensated orders and classify their spin textures using spin-space-group symmetry. We identify materials whose generated candidates are all unconventional, either across a defined search space or after restricting the magnetic-cell size. Within the experimental benchmark, 68% of the prioritized materials are unconventional, compared with 9% of the remaining materials. Screening the Materials Project yields thousands of promising candidates for unconventional compensated magnetism. First-principles calculations for $\mathrm{VGe_3}$ and tetragonal $\mathrm{Fe_2SiO_4}$ connect these symmetry predictions to the energetics and spin textures of competing magnetic orders. In $\mathrm{VGe_3}$, a noncoplanar candidate permits mixed-wave spin polarization along a fixed axis without spin-orbit coupling, combining components that are odd and even under momentum reversal. This framework enables crystallography-guided searches for unconventional compensated magnets without first determining their magnetic ground states.

Synthetic data can scale training supervision when real-world data are limited, but noise and distribution mismatch can reduce its value. Existing synthetic data selection methods often emphasize fidelity or diversity rather than the learner's evolving needs. We propose FROST, an online framework that estimates synthetic-data utility through gradient feedback anchored in real training data. It calibrates batch utility against recent history to determine when filtering is needed and filters samples only in out-of-band batches to determine what to retain, without an external verifier or held-out validation set. Experiments on two public benchmarks for image classification and LLM fine-tuning for text-to-SQL show that FROST filters out around 20–30% of the synthetic data while improving real-task performance compared with training on the full synthetic data pool. We further apply FROST during training in a large-scale industrial ads re-ranking system, achieving significant performance gains over a highly optimized production baseline, demonstrating its effectiveness and generalizability.

We give a constructive numerical treatment of Monge's optimal transport problem in 1D, 2D, 3D, following Evans-Gangbo: the map is built from a Kantorovich potential $u$ and density $a$ solving $-\mathrm{div}(a\nabla u) = f^+-f^-$, via $\dot T = -a\nabla u(T)/[(1-t)f^+(T)+tf^-(T)]$, $T(0,x)=x$. In 1D this gives a closed form. In 2D–left unfinished in an earlier report–we obtain $(u,a)$ via a convex Beckmann flow problem (second-order-cone programming), cross-checked against a dual LP, then build the map by direct integration or a ray-based 1D "clock" ODE with explicit density per ray. Validation uses strong duality, mass conservation, and vanishing-regularization limits, on a translation-invariant case and two 2D examples (a radial cone, a checkerboard) checked by symmetry; both show boundary-concentrated mass and non-unique flux where supports touch. We extend to 3D (theory unchanged), matching the 2D answer to six digits and reproducing the same phenomena on a radial ball, exposing a limit of the 2D visualization style. Movies illustrating the transport maps are available as ancillary files on the arXiv page of this submission.

With the help of a summation formula for $q$-series and the creative microscoping method, we shall establish two parametric $q$-supercongruences. They are both modulo the third power of a cyclotomic polynomial. When $q\to1$, one of them is able to engender the following conclusion: for any prime $p\equiv2\pmod{3}$ and any nonnegative integer $s$ subject to $ s\leq (p-2)/3$, \[\sum_{k=s}^{(p+1)/3+s}(6k-1)\frac{(-\frac{1}{3})_{k-s}(-\frac{1}{3})_{k+s}(-\frac{1}{3})_{k}}{(k-s)!(k+s)!k!} \equiv 0\pmod{p^3}.\]

Underwater 3D reconstruction is critical for marine exploration, ecological monitoring, and subsea infrastructure inspection, yet remains challenging at large scale due to light attenuation, scattering, and limited capture coverage. While 3D Gaussian Splatting (3DGS) enables high-quality real-time rendering, its application to large underwater scenes is constrained by high memory consumption and inefficient optimization over extensive areas. We propose OceanXL, a fast and scalable 3DGS-based framework for large-scale underwater reconstruction. OceanXL adopts a divide-and-conquer strategy, partitioning scenes into spatially coherent blocks to enable efficient optimization while preserving global geometric consistency. We further introduce an adaptive pruning scheme tailored to underwater conditions that removes redundant primitives, producing compact representations without sacrificing visual fidelity. Together, these components improve training efficiency and rendering performance for large scenes. We also introduce a large-scale underwater dataset covering diverse marine environments. Experiments on five large-scale scenes demonstrate favorable scalability, compactness, and efficiency–quality trade-offs over large-scene baselines. Controlled comparisons on the small-scale SeaThru-NeRF dataset further show competitive reconstruction quality with substantially smaller model sizes than underwater-specific methods.

Understanding the neurobiological mechanisms underlying self-referential cognition and baseline self-consciousness remains a fundamental challenge in computational neuroscience. In this work, we propose a large-scale computational model incorporating a 10,000-neuron spiking neural network (SNN) based on Izhikevich dynamics. The network is structured into two interacting subsystems: a sensory processing layer (5,000 regular-spiking cortical neurons) and an endogenous Default Mode Network (DMN) pacemaker subsystem (5,000 intrinsically bursting neurons). The DMN layer is modulated by continuous tonic currents reflecting ascending brainstem neuromodulation, maintaining intrinsic, autonomous bioelectric rhythms independent of external sensory input. To represent top-down cognitive modulation, synaptic weights are hierarchically structured such that DMN-to-network projections exceed sensory-level connections. Through numerical simulations using a modified two-step Euler integration scheme, we demonstrate how endogenous pacemaker activity interacts with transient external sensory perturbations, providing an elementary mathematical framework for the emergence of a persistent, self-sustaining neural representation of "Self".

Semantic IDs (SIDs) encode each catalog item as a short token sequence, enabling generative recommenders to predict the next item autoregressively. Reasoning-enhanced variants, an increasingly common extension, first generate a textual trace and then decode a next-item SID by beam search. Such recommenders are commonly trained with group-relative policy optimization under an exact-match SID reward, which is sparse in large catalogs. Two failure modes follow. When all rollouts in a group miss the target, the group yields zero advantage and no learning signal. Rollouts sharing the same SID reward receive identical advantages, however much their traces differ. In both cases the reward reflects only the decoded SID, never the reasoning that produced it. This creates a credit-assignment gap. We address this gap with retrieval-grounded query attribution. Each trace is structured into a history summary, a set of interest hypotheses, and a final SID. A frozen retriever executes every hypothesis as a catalog query, so that each hypothesis becomes independently verifiable rather than judged only through the final SID. A rollout is rewarded when any of its queries retrieves the target within the \mbox{top-$K$}, and per-query hit indicators localize that reward to individual hypotheses. Credit is thus assigned at the span level: only hypotheses that individually hit receive positive retrieval advantage, while the retrieval channel never updates the final SID span. Rollouts that share a SID reward can therefore receive different updates. Across experiments on three Amazon Reviews datasets, this yields consistent improvements in SID recommendation. On Video Games, an oracle analysis further reveals the potential of interest-conditioned SID decoding: selecting the target-relevant query among generated interests improves both recall and ranking.

By the paper \cite{Pet2009upper}, \cite{LiNan2026}, we know that there exists $C(n)>0$, for any complete compact or non-compact Riemannian n manifold $M$ with non-negative sectional curvature, without boundary and any $R>0$, \begin{equation} R^{2-n}\int_{B(p,R)} Scal \le C(n). \end{equation} Based on this result, we will prove that (1) There exists constant $C(n)$. If $M$ is a complete, n dimensional, non-compact Riemannian manifold with non-negative sectional curvature, then for any $p\in M$, $R>0$, \begin{equation} R^{2-n}\int_{B(p,R)} Scal \ dvol \le C(n)(1-v(M)), \end{equation} where $Scal$ is the scalar curvature and $v(M)=\lim\limits_{R\to \infty} \frac{volB(p,R)}{volB(0,R)}$ is the asymptotic volume ratio. (2) There exists constant $C(n)$. If $M$ is a complete, n dimensional Riemannian manifold with sectional curvature $\ge 1$, then \begin{equation} \int_M (Scal-n(n-1)) \ dvol \le C(n)(1-\frac{vol(M)}{vol(S^n(1))}). \end{equation}

This paper presents a Solidity, Hardhat, React, MetaMask, and ethers.js prototype for hospital ethics committee voting. Role controls, case-state checks, duplicate vote controls, and a receipt hash support public audit and transaction inclusion verification. Because vote events expose wallet addresses and vote values, the design provides pseudonymous auditability, not anonymous or secret-ballot voting; the receipt is neither receipt-free nor coercion-resistant. Evaluation reports 22 passing functional tests and local Hardhat gas use, including 284,137 gas per vote. A 12-participant simulation used assumed probabilities and is not human-subject evidence. Residual risks include multiple wallets, administrator or frontend compromise, credential reassignment, front-running, denial of service, and untested adversarial paths. Confidential deployment requires governed enrollment, encrypted ballots, independent audit, adversarial testing, reproducible benchmarks, and a real user study.

Let $(X,\rho,\mu)$ be a space of homogeneous type satisfying $\mu(X) =\infty$, the doubling property and the reverse doubling condition. Let $L$ be a nonnegative self-adjoint operator on $L^2(X)$ whose heat kernel has a Gaussian upper bound. First, we study the preduals of weighted Bourgain-Morrey spaces on $X$ and their properties, such as lattice property and completeness. Then we introduce the weighted homogeneous Besov-Triebel-Lizorkin block spaces. It is shown that they are the preduals of weighted homogeneous Bourgain-Morrey Besov-Triebel-Lizorkin spaces associated with the operator $L$. Consequently, these block spaces are independent of the choice of partitions of unity.

The observed cosmological constant is set by the deepest infrared scale of the present universe rather than by the microscopic scales that contribute to the vacuum energy. We show that this emerges as the quantum fluctuations of the gravitational field are progressively taken into account. Considering the Einstein-Hilbert+$R^2$ truncation of gravity, we find that in a wide region of the parameter space the theory develops instabilities that give rise to large amplitude quantum fluctuations which govern the renormalization group flow. Once the flow enters this region, it remains there all the way towards the infrared and progressively loses memory of its boundary conditions. The running cosmological constant $\Lambda_k$ ($k$ is the running scale) is driven towards the universal infrared scaling $\Lambda_k\sim \lambda_{_{\rm IR}}\,k^2$, with $\lambda_{_{\rm IR}}= \frac{1}{2(\pi-2)}$. At the deepest infrared scale $k_{_{\rm IR}}\sim H_0$ (the current Hubble scale), the cosmological constant is then screened to its observed value $\Lambda_{_{\rm IR}}\sim H_0^2$.

Predicting dielectric and piezoelectric responses in structurally complex materials requires large simulation cells, for which density-functional perturbation theory (DFPT) calculations scale as $\mathcal{O}(N^4)$ and become computationally prohibitive. However, machine-learning approaches have remained challenging, limited either by the large amounts of expensive DFPT data required for direct regression or by the cubic cost of reconstruction. We here propose a machine-learning framework that learns the ionic response as an atom-indexed field: DART (Direct Atom-Resolved response-Tensor learning) learns these fields directly and achieves high accuracy from small training sets, whereas LARS (Linear-scaling Atom-Resolved Response Solver) reconstructs them from learned microscopic ingredients through sparse linear solves and requires no DFPT labels for the ionic dielectric or piezoelectric tensors. Using DART, we extrapolate to 314 stackings of AlN/ScN superlattices absent from training and identify a high-response polar candidate, for which DFPT gives a laterally clamped piezoelectric strain coefficient $d_{33,f}=14.511$ pC/N and $k_t^2=20.46\%$, exceeding the ordered 1AlN/1ScN reference by 63% and 66%.

Branch model predictive control optimizes multiple future trajectories coupled through shared decisions, with computational demands increasing as the number of scenarios and prediction horizon grow. We present a GPU-accelerated direct linear solver for branch MPC formulations in which all trajectories share a single root decision node and evolve independently thereafter. By operating at the linear-algebra level, the solver provides a reusable backend for multiple optimization algorithms whose reduced systems have the required symmetric positive-definite structure. The solver exploits two levels of parallelism: across scenarios and along each prediction horizon. A tailored variable ordering enables horizon-parallel Cholesky factorization while preserving a single root-tail coupling block per scenario in the factor. Numerical experiments demonstrate substantial speedups over state-of-the-art sparse direct solvers, achieving factorization speedups of up to 6.0$\times$ over cuDSS and 27.6$\times$ over eight-thread PARDISO, with triangular solve speedups of up to 3.5$\times$ and 15.8$\times$, respectively.

Under Gaussian sampling, removing location and scale turns the sample into a direction that is uniformly distributed on a sphere of dimension $n-2$, and sample skewness and kurtosis become a cubic-quartic polynomial image of spherical measure. We use this geometry to obtain exact finite-sample results for the Jarque-Bera statistic $\mathrm{JB}_n$. First, for every $n\ge 3$ the largest attainable value of $\mathrm{JB}_n$ is $n\{(n-2)^4+4(n-1)^2\}/\{24(n-1)^2\}$, attained by a single-outlier sample, so the asymptotic $10\%$ and $5\%$ tests have size zero for $n\le 6$ and the $1\%$ test for $n\le 7$; near this maximum the density of $\mathrm{JB}_n$ behaves like an explicit multiple of $(J_n^+-x)^{(n-4)/2}$. Second, passing from residual coordinates to power sums writes the joint density of skewness and kurtosis as an integral of the reciprocal square root of a polynomial discriminant; for every $n\ge 5$ the innermost integral runs over a single interval and is a Lauricella $F_D$ period. Third, this yields explicit laws: an arcsine law for $n=3$, an algebraic joint density for $n=4$, and a single complete elliptic integral, equivalently ${}_2F_1(\tfrac12,\tfrac12;1;\cdot)$, for $n=5$. Residual-coordinate collisions generate the discriminant singularities of the joint law, while stationary points of $\mathrm{JB}_n$ on the residual sphere govern the singularities of its one-dimensional density.

A routing node on the Lightning Network holds its liquidity in separate channels, so a payment can fail at a channel whose outbound balance is exhausted while the node's other channels still hold balance. Pooling the channels into one reserve fixes this only if the node's draws across all channels stay within the reserve: a global invariant that each counterparty must enforce from its own channel, with no shared counter that off-chain draws can update. A node must therefore split the reserve into per-channel quotas in advance or coordinate every draw with every counterparty. On three Lightning snapshots the advance split forfeits $16$ to $67\%$ of the pooling gain over unpooled channels, and the loss grows with channel count. Sluice recovers $19$ to $65\%$ of that loss with a nested reservation: each channel keeps an exclusive base that its counterparty checks alone, and the rest is a shared overflow drawn on with certificates from a capacity-weighted quorum of the node's counterparties. Two conflicting certificates share an honest signer, so over-drawing is prevented without a slashable stake, under a stated bound on the capacity the node controls in the signing set. Sluice loses at most $1.3$ points of payment success to coordination where deployed coin movers lose up to $10.3$, and improves them in eleven of twelve cells when stacked on them. Re-creating every base output each epoch costs $2.8$ to $5.1$ times Lightning's on-chain bytes; re-creating only those that overflowed costs $0.5$ to $1.2$ times and keeps part of the gain.

Predicting heterogeneous causal effects supports informed decision-making, yet suitable inductive biases vary with treatment overlap, covariate imbalance, prognostic structure, and sample size. The Geometry-Diverse Anchor–Correction Expert Ensemble (GeoACE) addresses this variation through five complementary experts whose predictions are combined using weights estimated from validation data. Compared with ensemble designs that derive diversity primarily from sample or feature subsets, reweighting, or feature transformations, its central contribution is the coordinated design of training objectives and anchor geometries that target different prognostic and overlap structures. Following validation, each expert is refitted on the combined training and validation data, while retaining its distinctive structure. The shared architecture couples a structured outcome anchor with a learned neural correction. ACE serves as the reference estimator, while OW-ACE emphasizes well-supported covariate regions through overlap-weighted factual training. G$\Phi$-ACE constructs a global outcome-guided geometry from pooled prognostic associations, whereas A$\Phi$-ACE incorporates arm-specific outcome associations and penalties for treatment-group imbalance. O$\Phi$-ACE complements these approaches with an outcome-free, overlap-aware projection that replaces the anchor input with a lower-dimensional representation. GeoACE is evaluated against 11 comparators on eight benchmark protocols, alongside comparisons of weighting rules and aggregation controls. It achieves the lowest mean $\sqrt{\mathrm{PEHE}}$ and absolute average treatment effect error on all three IHDP protocols and closely approaches the best comparator on NEWS, although performance varies across the remaining benchmarks. The aggregation analyses show similar observed performance for inverse-DR weighting and equal averaging.

Generative recommendation reformulates item retrieval as sequence generation, allowing a unified model to directly generate the next item from a user's interaction history. Semantic IDs further make this paradigm effective and scalable by representing each item as discrete codes, enabling knowledge sharing among semantically related items. Recent studies introduce explicit reasoning before Semantic-ID generation, helping models summarize user interests and infer possible preference transitions. However, reasoning is not inherently beneficial: Inaccurate or uninformative reasoning may mislead subsequent item generation and ultimately degrade recommendation performance. This raises a central challenge: how can a recommender select and learn effective reasoning traces and progressively evolve toward better reasoning from its own generations? In this work, we propose Evo-Rec, a three-stage framework for learning better reasoning and further enhancing it through reinforcement learning. First, we align Semantic IDs with their textual and behavioral contexts, enabling the model to understand and generate item identifiers. Second, we sample multiple candidate reasoning traces and retain those that improve the prediction of the ground-truth item, providing a stronger reasoning initialization through supervised fine-tuning. Third, we further optimize the reasoning policy through reinforcement learning with catalog-constrained item generation and ranking-aware recommendation feedback. Experiments on three Amazon Review benchmarks show that Evo-Rec consistently outperforms discriminative, generative, and reasoning-enhanced recommenders across all evaluation metrics. These results demonstrate the effectiveness of our framework in learning better reasoning for SID-based generative recommendation.

We develop and analyse a mechanistic ODE model of within-host HIV dynamics that includes uninfected CD4$^+$ T cells, latently and productively infected cells, free virions, and cytotoxic immune effectors. Antiretroviral therapy is described by two time-dependent efficacy functions that separately reduce new infections and virion production. We show that solutions remain non-negative and bounded, identify the infection-free and endemic equilibria, and derive the basic reproduction number together with the local stability condition for the infection-free state. We also obtain treatment-dependent suppression thresholds and an analytical estimate of when this threshold is crossed as treatment efficacy declines. Numerical simulations consider untreated infection, ART initiation, periodic variation in efficacy, and progressive loss of treatment effect. Without therapy, the model approaches a state of persistent infection. ART reduces viral load and promotes recovery of the CD4$^+$ T-cell population, while the latent reservoir persists under the parameter set considered. Periodic changes in treatment efficacy produce sustained forced oscillations. These results connect the analytical threshold conditions with the treatment-dependent dynamics of the model and provide a basis for future calibration, sensitivity analysis, uncertainty quantification, and extensions informed by clinical data.

Nanoparticle properties are governed by their surfaces, yet their atomic-scale surface structures remain challenging to predict. Here, we report that bimetallic nanoparticles can form a thermodynamically controlled "shell-dimer" architecture in which one metal forms an overlayer only a few atomic layers thick on another. Using Au-Rh as a model system, atomic-resolution imaging and molecular dynamics show that an ultrathin Au overlayer forms on Rh, stabilized by competition among surface, interfacial, and strain energies. Anisotropic strain limits its growth to the subnanometer scale, and changes in surface chemistry can destabilize the overlayer altogether. Across a range of bimetallic nanoparticles, we found that overlayer formation is associated with elemental immiscibility and lattice mismatch. These findings provide a basis for understanding and controlling surface structures in multimetallic nanoparticles.

Quarkonium production has long been considered as one of the golden probes to study the quark-gluon plasma (QGP). In fact, the early production of heavy quarks (ccbar) and (bbbar) makes quarkonia an ideal tool to investigate the evolution of the hot and dense medium produced in ultra-relativistic heavy-ion collisions. In such a medium, quarkonium production is expected to be suppressed due to the screening of the binding potential between the heavy quark and antiquark and/or through in-medium interactions. Moreover, at LHC energies the recombination of uncorrelated charm quarks pairs, namely regeneration, was found to significantly affect charmonium observables, in contraposition to the suppression mechanism. In addition, measurements in smaller collision systems as p-Pb have highlighted the possibility to observe QGP-like effects. In this context, the study of charmonium production in intermediate collision systems, as light-ion collisions, becomes more and more interesting, representing an ideal test ground for the state-of-the art theoretical models. In this contribution the new measurements of charmonium production will be shown using the light-ion collisions data collected for the first time at the LHC in 2025 (oxygen-oxygen (OO), proton-oxygen (pO)). The results will be shown exploiting the forward ALICE rapidity coverage (2.5 < y < 4). Finally, the measurements will be compared with the existing theoretical models.

Many inverse problems in physics infer an unobserved field from measurements connected to it through a governing equation. We combine Fisher information with a differentiable solver to quantify uncertainty in both the inferred field and its predicted observables, using a local Gaussian approximation without sampling the full parameter posterior. Applied to nuclear optical potentials, the method distinguishes features constrained by scattering measurements from those whose uncertainty remains set by the prior. Our flexible model improves agreement with held-out cross sections and achieves nearly nominal coverage with narrower predictive intervals than an established Bayesian optical model.

In this paper, we prove the equivalence between coerciveness and the existence of a solution to the generalised Monge-Ampère equations for right-Noetherian polynomials satisfying several assumptions. This includes the case of strictly interlacing polynomials. In particular, we have generalised Chu-Lee's result to supercritical but not hypercritical LYZ equations.

We show that for a 1-category C, the ({\omega}, k)-polygraphic homology of C for any k {\geq} 1, that is taken with cofibrant resolutions in strict ({\omega}, k)- categories, does not depend on k and is canonically isomorphic to the homology of the classifying space of C. When C is a groupoid, we also show this for k = 0. In particular, this means that the classical homology of groups can be obtained by taking cofibrant resolutions in strict {\omega}-groupoids. In order to show these results, we develop the theory of discrete Conduché fibrations in the category of strict ({\omega}, k)-categories, building on previous work by the first-named author.

Zero Forcing (ZF) and Power Dominating Set (PDS) mark vertices in a graph based on a common forcing process starting from a problem-specific set of initially marked vertices. ZF initially marks the selected vertices while PDS additionally marks their neighbors. In the forcing process, a marked vertex with only one unmarked neighbor may force that neighbor which then becomes marked, too. A solution marks the entire graph by exhaustive application of this rule. One variant generalizes the forcing threshold; vertices may force when they have a fixed number of $k$ unmarked neighbors. Another variant limits propagation to a fixed number of rounds. We classify the parameterized complexity of the problem variants obtained by combining these choices of initialization, round limit and forcing threshold. We show that with appropriate choices, these variants range in parameterized complexity from fixed-parameter tractable to complete for every even layer $W[2\ell]$ of the $W$-hierarchy, and up to $W[P]$-complete. Our results demonstrate that small changes in any one of these three dimensions can lead to a sharp change in problem complexity.

Standard-cell libraries form the critical interface between transistor-level circuit design and automated digital design flows, yet their design and characterization are rarely covered in depth in digital VLSI courses. This paper presents VLSI 5, a graduate-level course at ETH Zurich in which students design, lay out, characterize, and integrate their own standard cells using IHP's open-source 130nm SG13G2 process design kit. In the first course offering, 19 students developed EZ130 8T, an open-source eight-track standard-cell library comprising 220 cells. We evaluate successive library versions using eight synthesis benchmarks, across which EZ130 8T achieves significant area reduction compared to the SG13G2 library at competitive timing. For a postlayout 8x8 matrix-vector multiplier, EZ130 8T reduces cell area by 35% and energy by 39% while maintaining virtually the same clock period. The library has already been adopted in other ETH Zurich VLSI courses, enabling the creation of chips entirely designed by students down to the transistor level.

General-purpose vision-language models (VLMs) bring broad knowledge and spatial reasoning to robot manipulation, yet existing systems either use them indirectly, to predict constraints or write programs, or give them a view of the scene rather than a world in which to act. We present World Action Agent (WAA), a multi-agent harness through which VLMs pilot robots with basic tools, making every decision within a visual action workspace. The workspace has three properties. Contact views, selected automatically from the scene geometry, present the scene around the current interaction. Action rehearsal turns each action into an editable proposal that the agent, alone or through an Imagination Agent, previews and revises against planning feedback before execution. In-view correction closes the loop between observation, rehearsal, and low-level execution, letting the agent remove residual offsets in the view where it observes them. Through the same workspace, WAA acquires embodied procedural knowledge in two ways: it evolves multimodal skills from expert videos and human teaching under evidence-based review and consults them through a Skill Agent, and its interaction traces train smaller VLMs to pilot the same harness. On LIBERO-Pro, WAA with skills evolved only from LIBERO-90 reaches a state-of-the-art 75.6% average success, outperforming end-to-end VLAs, code-as-policy agents, and a visual-harness baseline with the same backbone; the same skills remain effective on robosuite without further learning. Fine-tuning Qwen3.5-9B on harness traces raises its out-of-domain success from 1.7% to 43.3%.

Textured Gaussians improve local appearance capacity, but assigning the same texture resolution to every primitive wastes storage on low-detail or weakly visible regions. We introduce AdaTex4D, an adaptive texture-capacity module for deformation-based 4D Gaussian Splatting. Each Gaussian carries packed RGBA triplanes whose two axes grow independently according to visibility normalized screen-space gradients and deformed local scales. Experiments on N3DV and PanopticSports show that AdaTex4D reduces texture storage by more than half while preserving reconstruction quality. Under fixed memory budgets, adaptive allocation also improves quality over uniform texture assignment and reduces overall model and peak memory. These results show that dynamic, anisotropic texture allocation provides a more efficient way to distribute local appearance capacity in 4D Gaussian representations.

Formal verification tools commonly rely on SMT solvers to automatically reason about programs, leveraging a range of logical theories, e.g., linear integer arithmetic, arrays, or strings, to encode program constructs and verification conditions. Despite recent advances, such solvers still struggle when reasoning about recursive data structures such as lists, which are pervasive in modern functional languages. Additionally, lists are commonly used in conjunction with higher-order combinators to, e.g., generically apply a function to all elements of the list. In this work, we provide first-class support for reasoning about lists within SMT solvers. We focus on lists of arbitrary size that, following the map-reduce paradigm, can be manipulated exclusively through a set of abstract combinators. To this end, we introduce DueList, an abstraction-refinement approach geared towards list reasoning, which we implement on top of off-the-shelf SMT solvers. To evaluate the efficiency of our approach, we assemble a diverse set of 752 benchmarks curated from previous works and real-world programs, and compare DueList against state-of-the-art solvers such as Z3 and CVC5. Our experimental evaluation shows that DueList extends reasoning facilities of existing solvers, allowing to conclude about the (un)satisfiability of a larger range of problems, while outperforming existing solvers in the vast majority of previously supported cases.

Many iterative algorithms rely on bootstrapping. A variable is updated using a second, frozen copy as a target, which is periodically replaced with the updated variable. Majorize-minimize and inexact proximal-point methods share this structure, as does temporal-difference (TD) learning. However, existing convergence guarantees for scenarios that combine sampled updates with targets refreshed only every $K$ steps rely on the specific structure of the update, such as linear approximation or gradient-based inner steps, and on uniformly bounded sampling error. We instead model the sampled update as a stochastic operator on the parameter space, which reduces the analysis to a contraction argument that needs no gradient structure and allows the sampling error to grow with the iterates. Within this framework, we derive a finite-time bound for i.i.d. samples and any target-update period $K$. We show that the iterates converge geometrically in root mean square to a ball around the fixed point, provided the sensitivity to the frozen target is smaller than the contraction slack of the inner map. Existing deterministic frozen-target contraction and stochastic-gradient-type bounds follow as special cases of our framework, and simulations of TD learning reproduce the predicted contraction rate and scaling of the error floor with the step size.

Fine-tuning large language models on new objectives can improve helpfulness, instruction following, or domain-specific performance, but it can also induce regressions on safety-critical prompts. Existing safety-preserving fine-tuning methods typically control average safety loss or use weighted auxiliary penalties, which can obscure rare but severe failures. We propose a chance-constrained formulation for safety-preserving fine-tuning that limits the fraction of safety examples whose degradation relative to a reference model exceeds a prescribed threshold. Because the resulting empirical chance constraint contains a discontinuous indicator, we introduce a differentiable majorization of the violation rate, yielding a tractable conservative constraint. We then develop a constraint-aware gradient descent method that treats the majorized constraint as a safe set in parameter space and minimally modifies the fine-tuning direction to preserve feasibility. The resulting update admits a closed form and produces a tail-aware safety correction that emphasizes examples near or above the degradation threshold. We conduct an extensive set of experiments on harmful fine-tuning across three different tasks and three models and show that our approach consistently outperforms the baselines that exist in the literature. These results suggest that safety preservation in LLM fine-tuning is better viewed as a reliability-constrained optimization problem than as average-risk regularization.