Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Monday 14 September

Artificial Intelligence

Challenge-Aware Knowledge Recruitment with Meta-Agent Supervision for Medical Multi-Agent Reasoning

Large language models perform strongly on standard medical QA but degrade on harder reasoning tasks, where multi-agent debate often fails to beat strong single-model prompting. We argue that a key limitation is that existing systems select experts and retrieve knowledge from the question alone, rather than from gaps that emerge during reasoning. We propose challenge-aware adaptive recruitment: an external CHALLENGER inspects the evolving reasoning trace and identifies specific factual or expertise gaps. These signals trigger targeted knowledge retrieval or mid-discussion specialist recruitment. On the MEDAGENTSBENCH hard subset (862 questions across nine biomedical datasets), upfront retrieval provides little benefit, while challenge-aware in-between retrieval performs better; on a multi-specialist backbone it improves accuracy on 8 of 9 datasets and adds 6 percentage points over Challenger Only. Ablations indicate benefits from both retrieval timing and gap-specific query signals, and a live web-search experiment shows that the timing advantage persists across retrieval backends. For specialist recruitment, self-triggered adaptation is unreliable, whereas Challenger-triggered recruitment raises GPT-4.1-mini MedQA accuracy from 46% to 54%, with positive direction on all tested generalization datasets. Naively combining retrieval and specialist recruitment degrades performance, highlighting the need to coordinate adaptive intervention channels. These results show that when and why external resources are recruited is a key design choice in multi-agent medical reasoning.
Read the full abstract

Artificial Intelligence

Uncertainty-Aware Resource Orchestration for Agentic AI Systems

Agentic artificial intelligence (AI) systems powered by large language models (LLMs) execute as autonomous multi-step pipelines that dynamically invoke tools, integrate observations into memory, and iterate until a task is resolved or a resource budget is exhausted. Unlike classical workflows with fixed graphs, the aggregate compute, bandwidth, and end-to-end latency demands of such pipelines are unknown at dispatch time: they emerge at runtime from intermediate observations and stochastic token sampling, yielding heavy-tailed, multi-modal random variables. This unpredictability forces infrastructure orchestrators into a dilemma: under-provisioning risks mid-execution throttling and deadline violations, while over-provisioning wastes scarce edge or cloud capacity. Existing schedulers require a fully known pipeline graph, while edge resource-management methods rely on parametric or stationary demand models, and neither fits the unknown, non-stationary profiles of agentic pipelines. We formalize this setting as the uncertainty-aware agentic orchestration problem, where the goal is to reserve, at dispatch time, compute and bandwidth allocations that cover actual pipeline demands, and to verify that the pipeline can meet its deadline, each with a prescribed probability. We then propose Conformal Reservation for Orchestration (CRO), an online algorithm that translates a sliding window of past execution traces into dispatch-time resource allocations with finite-sample, distribution-free coverage guarantees via conformal prediction. Simulations over heterogeneous task mixes with multi-modal and heavy-tailed profiles show that CRO is the only evaluated method that simultaneously satisfies target probabilistic coverage across all risk levels while, among coverage-feasible methods, attaining the lowest over-reservation, roughly halving the resource waste of the Chebyshev-bound baseline.
Read the full abstract

Computation and Language

Translating in style: An open-weight alternative to GPT-4 for culturally aligned machine translation

Machine translation systems often preserve meaning but systematically flatten linguistic style, a failure that is most acute when translating into non-Western languages. The RASTA framework of Havaldar et al. (2025a) addresses this through retrieval-augmented stylistic alignment, but relies on GPT-4, a closed, proprietary model. We investigate whether an open-weight translation specialist, Hunyuan-MT-7B, can serve as an open-weight alternative. We evaluate three variants of the model, the base model, a LoRA supervised fine-tune, and a GRPO-steered policy. We do this across three style dimensions (formality, politeness, and intimacy) and twelve directed language pairs, measuring style alignment with the correlation-based metric $\mathcal{A}$. We find that fine-tuning does not reliably improve alignment: the three variants are statistically indistinguishable in aggregate. The open backbone does not match RASTA-GPT-4 overall, but the shortfall is concentrated in formality and largely closes for intimacy and Chinese directions, where the open model matches or surpasses GPT-4 on individual pairs. An accessible RASTA pipeline built on an open model is thus already viable for parts of the problem.
Read the full abstract

General Mathematics

Tensor Degenerations for Matrix Multiplication

Let ${\mathbb K}$ be an arbitrary field. We study tensor degenerations for square matrix multiplication over the field ${\mathbb K}$, and mainly prove for any integer $n\ge5$ the uniform dyadic estimate \[ \underline{\mathrm R}_{\mathbb K}({\mathsf M}_{2^n}) \le 2^{2n+1}\] which implies that $\omega(\mathbb K)\le 2+1/n$, where the border rank $\underline{\mathrm R}_{\mathbb K}(T)$ is the least $r$ for which the scalar extension of $T$ to an algebraic closure of $\mathbb K$ lies in the Zariski closure of the tensors of rank at most $r$, and $\omega(\mathbb K)$ is the matrix multiplication exponent over the field ${\mathbb K}$. This improves the current best record $\omega(\mathbb K)<2.371177$ greatly, and answers affirmatively the long-standing open question whether the value $\omega({\mathbb K})$ is exactly $2$. The implementation disclosure and appendices in this paper give fixed source maps, interpolation data, terminal incidence ledgers, and deterministic check counts for independent reconstruction.
Read the full abstract

Rings and Algebras

Products of Boolean Clones up to Minion Homomorphisms

We study the minion homomorphism poset of multisorted Boolean clones. Our main result completely determines the subposet induced by products of Boolean clones: every inequality between prod- ucts is induced by inequalities between their Boolean factors. Furthermore, we determine the minion cores of all products of Boolean clones, thereby identifying canonical representatives of their minion homomorphic equivalence classes.
Read the full abstract

Neurons and Cognition

Interpretable Memory Models for Spaced Repetition

Spaced repetition software schedules reviews with a memory model fit to review logs. Accuracy on a test set is not sufficient evidence of quality since available data are produced by existing schedulers, and new solutions must extrapolate beyond them. A model also needs a simple mechanistic interpretation. We present SBD, a model that is more interpretable and 80% smaller than the current state of the art at nearly the same accuracy.
Read the full abstract

Artificial Intelligence

Accelerating Floating-Point Satisfiability Solving via Gradient Normalization

Satisfiability Modulo Theories (SMT) solvers are foundational to software verification, program analysis, and compiler testing, particularly over the theory of Quantifier-Free Floating-Point (QF_FP). While recent optimization-based SMT solvers have successfully applied gradient descent to continuous relaxations of logical formulas, they are fundamentally bottlenecked by gradient domination, a phenomenon where a small subset of difficult clauses hijacks the optimization trajectory, preventing the solver from satisfying the broader formula and trapping it in local minima. To overcome this, we present GradSAT, a novel framework that bridges optimization-based SMT solving with Multi-Task Learning (MTL). GradSAT reformulates the constraint satisfaction process by treating each SMT clause as an independent MTL task. By applying dynamic gradient normalization (GradNorm), GradSAT actively balances the gradient magnitudes across all clauses at runtime, systematically penalizing dominant gradients and accelerating lagging clauses to ensure uniform convergence. GradSAT implements this through a highly optimized, two-stage hybrid pipeline. First, a GPU-accelerated PyTorch backend leveraging symbolic compilation and operator fusion navigates the continuous relaxation to a high-quality basin. Second, the candidate assignment is handed off to a bit-precise local search engine to rapidly resolve the exact, rigorous assignment. By stabilizing the continuous search dynamics, GradSAT mitigates the brittleness of prior gradient-based solvers and provides a robust, highly parallelizable architecture for complex constraint solving.
Read the full abstract

Algebraic Geometry

A toric Calabi-Yau counterexample to Schimpf's connected stable pairs pole conjecture

Schimpf's connected reformulation of Pandharipande's pole conjecture for stable pairs predicts the following restriction: in class \(\beta\), connected descendent coefficients have nonzero \(p\)-poles only where \(-p\) is an \(m\)-th root of unity with \(1\leq m\leq\divis(\beta)\). We give a toric counterexample. The threefold is a smooth quasi-projective toric Calabi–Yau threefold whose compact torus-invariant curves form a chain \(C_1\cup C_2\). For the primitive class \(\beta=2[C_1]+[C_2]\), the coefficient of \(Q^\beta z^3\) in the connected series for the insertion \(\ch_z([v]_T)\), where \([v]_T\) is the equivariant class of \(v=C_1\cap C_2\), has Laurent expansion \[ {\frac{9}{4}}\frac{1}{p-1}+O(1) \] after restricting to the one-parameter subtorus with tangent weights \((1,2,-3)\) at \(v\). Since \(\divis(\beta)=1\), Schimpf's condition allows only the nonzero pole \(p=-1\). The pole at \(p=1\) is therefore forbidden.
Read the full abstract

Tissues and Organs

Brain scaling and diversity in musteloids

The diversity of brain size and organization across mammals reflects how evolutionary and ecological pressures have shaped neural architecture through time. Yet, how these selective pressures translated into measurable variation in the proportions and overall shape of the brain across species remains poorly understood. In this study, we used high-resolution virtual endocasts from 41 musteloid species (e.g. otters, skunks, raccoons) to quantify variation in endocranial composition and shape. We analyzed both the complete endocast and its principal subdivisions (olfactory bulbs, cerebrum, and hindbrain) using a combination of compositional data analysis and 3D geometric morphometrics. Phylogenetically informed regressions (PGLS) and model comparisons were used to assess the effects of endocranial size, locomotion, diet, and taxonomy on both the regional proportions and the overall shape of the endocranial cast. Musteloid brain organization is mostly driven by size-dependent scaling: larger brains have enlarged cerebra, reduced olfactory and hindbrain regions, and rounder endocasts. However, some lineages markedly deviate from this largely conserved allometry: otters (Lutrinae) exhibit proportionally enlarged cerebra and strongly reduced olfactory bulbs, whereas skunks (Mephitidae) show the reverse pattern, emphasizing enhanced olfactory investment. Feeding ecology also contributes to variation in brain organization, as piscivores show disproportionately smaller olfactory bulbs compared with other feeding guilds. In contrast, locomotor mode has little explanatory power for either endocranial composition or shape. Brain evolution in musteloids is therefore not uniform: a common scaling pattern dominates, but lineage-and diet-specific adjustments modify how brain regions are proportioned.
Read the full abstract

All authors Margot Michaud, Arianna Rose Harrington, Sandrine Ladeveze, Eloy Gálvez-López, Valentin Fischer, Apolline Alfsen, Ornella Bertrand

Tissues and Organs

Rare Disease Classification in Video Capsule Endoscopy

Video capsule endoscopy offers a non-invasive way to examine the small bowel, however, the rare but clinically important findings that appear only in less than 1% of labelled frames, provide very less signal for detection. As a result, while models can achieve high overall performance, they almost completely miss these rare findings. We condition each frame's decision on evidence from the rest of that patient's video, not on the frame's own posterior alone, supplying information a single frozen frame can never contain. Building upon Expected Gain Arbitration, a per-frame, macro-F1-optimal correction over the confusion matrix, we further compose it with video-disease-presence conditional resulting in our decoding method Hierarchical Conditional-Expected-gain (HCE). We evaluate HCE on two independent VCE datasets, Kvasir-Capsule and GALAR, against a static temporal-smoothing baseline and an optimal-transport (OT) frame-level correction. HCE significantly improves rare-class recall over a temporal-smoothing baseline on GALAR (2.7–3.1x; 95% CI excludes zero in both scopes) and shows a similar, though not statistically confirmed, gain on Kvasir-Capsule (up to 3.4x). Against OT, HCE still improves rare-class recall in all four evaluation settings (+0.016 to +0.029), and additionally improves macro-F1 and MCC.
Read the full abstract

Computation and Language

Metric-Construction Coupling Inflates Measured Synthetic Dialect Recovery

Korean dialect corpora are available but not redistributable: weights may be released, while reproducing training and evaluation from the underlying data cannot be. We ask how much of that supervision synthetic data recovers, and whether that recovery can be measured independently of the synthesis pipeline. We contribute KoDialectBench, 1,000 items across five regions on three axes, released as identifier hashes and scoring code so users reconstruct the items from their own licensed copy. Recovery is strongly axis-dependent: our best synthetic arm reaches 91.2% of the real-data gain on region identification but 63.7% on comprehension. On generation the answer depends on the metric: the deployed marker lexicon reports 92.3% on dialectness and 119.3% on region match, the latter exceeding the real-data reference, whereas reference-based generation reaches 72.4%. We find the marker metrics' scoring inventory is entirely contained in the inventory our transformation rules can emit. We test the effect of construction access directly with an exact-form construction-disjoint arm that withholds 20% of marker types from the rules. At exactly matched training size (8,600 examples) it reduces dialectness recovery from 91.8% to 8.1% and region-match recovery from 101.9% to 25.6% on the held-out marker inventory, while the three pipeline-independent measurements do not fall at all. A complementary evaluator sweep defines metric-construction coverage (MCC) and finds measured dialectness recovery increasing monotonically as overlap rises from MCC=0 to MCC=1. Shared construction and evaluation inventories can therefore substantially inflate estimates of synthetic-data recovery.
Read the full abstract

Human-Computer Interaction

When the AI Leaves the Tailorshop: Measuring What an LLM Advisor Leaves Behind in Complex Problem Solving

Complex problem solving depends on acting effectively and understanding how a system works. AI advice may support these outcomes unequally. Two preregistered experiments compared participants managing a simulated clothing factory with and without an LLM advisor. Across studies, AI-supported participants reported greater confidence and understanding with less effort. In the first study (N=200), assistance increased company value but produced no detectable prediction-accuracy difference. After withdrawal, previously supported participants outperformed controls when decisions were scored against repeating previous choices, but not default settings. Within the AI-supported group, more frequent recommendation alterations predicted better unaided performance. In the second study (N=198), AI-supported participants went bankrupt less often and showed a small knowledge advantage in the registered analysis, largely associated with remaining solvent. More frequent recommendation alterations predicted higher knowledge within the AI-supported group. Applied HAI evaluation should assess users' understanding and independent capability alongside the performance achieved with AI support.
Read the full abstract

Analysis of PDEs

Global Strong Solutions for the Ternary Maxwell-Stefan Diffusion System

We prove global strong well-posedness for the normalized isothermal, isobaric ternary Maxwell-Stefan diffusion system on bounded $C^{2+\alpha_0}$ domains, for arbitrary positive constant binary frictions and compatible nonnegative initial data, without a smallness condition. For distinct frictions, trace–determinant variables produce two separated spectral roots satisfying scalar parabolic equations with antisymmetric quadratic coupling. Euler-Poisson-Darboux one-sided truncations, stabilized by the mixing entropy, yield boundary De Giorgi oscillation decay and positive-time Hölder regularity. The mixing-entropy Hessian then provides a local symmetrizer; conormal Schauder estimates give uniform $C^{1+a}$ control, and frozen maximal $L_p$-regularity yields a uniform terminal trace bound excluding finite-time breakdown. Repeated frictions reduce to triangular scalar parabolic systems. Every component with positive conserved mass is strictly positive for positive time.
Read the full abstract

Optimization and Control

Reachability and exact controllability of linear Stieltjes differential systems

We study reachability and exact controllability for systems driven by a nondecreasing left-continuous Stieltjes derivator whose measure may have continuous, singular-continuous and atomic parts, while flat intervals represent inactivity. We construct an evolution family and post-jump variation-of-constants formula, and characterize the reachable subspace as the essential span of input directions. For constant coefficients, the Kalman condition is necessary for every derivator and, under regressivity, sufficient when the derivator is continuous and nonconstant on a subinterval. If the measure is purely atomic with exactly $N$ positive atoms, the reachable subspace is the truncated Krylov space generated by $B,AB,\ldots,A^{N-1}B$; for a Kalman-controllable pair, the minimum atom count among purely atomic derivators equals the controllability index. At the threshold count, one vanishing mass with the others converging to positive finite limits causes Gramian degeneration and worst-case energy blow-up; at or above threshold, compact positive mass ranges yield uniform coercivity. A mixed double-integrator example illustrates the results.
Read the full abstract

Graphics

Hybrid Reconstruction of Admissible Spline Spaces from Locally Modified Unclamped Patches for Isogeometric Analysis

We propose a decoupled reconstruction framework that temporarily decomposes a NURBS representation into independent local Active Sections, enabling arbitrary local knot insertion, degree elevation, and basis modifications while preserving exact CAD geometry. However, these independent modifications inevitably violate inter-patch continuity, requiring a robust algebraic recovery of global admissibility. We introduce a novel reconstruction methodology that constructs a positive Hybrid Reconstruction Operator: anchor degrees of freedom are identified via QR pivoting, after which a sequence of linear programming problems with distance-based regularization generates strictly non-negative nullspace vectors, while a subsequent non-negative least-squares solve enforces a precise partition of unity. Crucially, to ensure computational efficiency and preserve locality, we implement a hierarchical pairwise condensation strategy that freezes columns already satisfying new interface constraints, confining the optimization exclusively to the active degrees of freedom at each merge step. The resulting hybrid basis is, by construction, strictly non-negative, spans the exact constrained nullspace, preserves partition of unity to machine precision, and reproduces the original geometry exactly. Numerical benchmarks, including a nonlinear diffusion problem on multi-patch curves with heterogeneous local polynomial degrees, confirm optimal convergence rates and demonstrate the framework's ability to seamlessly combine local geometric flexibility with globally consistent approximation spaces. By entirely decoupling local spline editing from the enforcement of continuity, this methodology provides a highly efficient, mathematically principled, and universally applicable tool for adaptive Isogeometric Analysis.
Read the full abstract

Machine Learning

Beyond the Graph: An Adaptive Meta-Learner Fuses Explainability, Weather, and Dynamics for Robust Bus ETA Prediction

Accurate bus Estimated Time of Arrival (ETA) prediction is vital for urban mobility, passenger satisfaction, and transit efficiency, yet existing models falter against nonlinear spatiotemporal dynamics, data sparsity, and factors such as weather. This paper proposes HYB(nm), an adaptive hybrid ensemble framework that dynamically fuses five complementary models - a historical baseline (MST-AV), periodical temporal pattern analysis (GDRN-DFT), Koopman Neural Operators for nonlinear dynamics (KOOP-NET), weather-integrated feature-engineered neural networks (FENN), and real-time graph convolutional networks (MGCN) - via a meta-learner attuned to real-time context. Evaluated on GPS and weather data from three Kolkata bus routes comprising more than 4,000 trips, the framework leverages the individual strengths of its components (for example, the low-latency explainability of MST-AV, the weather resilience of FENN, and the network-dynamics capture of MGCN) to deliver the superior robustness of HYB(2), state-of-the-art accuracy rivalling leading graph neural networks, and balanced trade-offs in stability and efficiency across prediction horizons and operating conditions. The extensible HYB(k) architecture equips transit agencies with flexible tools, ranging from economical single models to tailored high-fidelity hybrids, advancing predictive, equitable urban transport.
Read the full abstract

Computer Vision

Temporal-Attention Head Specialization During Video Diffusion Training

Video diffusion transformers depend on temporal attention to coordinate information across frames, yet nearly everything known about this mechanism comes from analyzing trained models, so when and where temporal-attention structure forms during training remains poorly characterized. Population averages can also hide it, since a few specializing heads and a diffusing majority cancel in the mean. We therefore conduct a checkpoint-resolved census of every temporal-attention head across nine Open-Sora STDiT training runs spanning three model scales (306M to 1.03B parameters), scoring each head with an entropy-normalized measure of cross-frame attention concentration (CFAC) under a preregistered change-point and effect-size selection rule. The census reveals the sparse picture that averages obscure. Aggregate CFAC is flat or decreasing in every run, while a small minority of heads, roughly 4–13% in full-grid runs, develops pronounced concentration. Across seeds, the reproducible signal is positional but block-level. Selected heads repeatedly arise in the first temporal block, whereas individual head coordinates do not reproduce once block membership is accounted for. Among the analyzed 760M selected heads, attention maps converge to a small repertoire of local frame-routing motifs, self-frame diagonals and adjacent-frame bands, even when the responsible coordinates differ across runs. Correlation and ablation analyses do not establish a causal link to generated video quality, and we bound our claims accordingly. Beyond this STDiT family, the study contributes a transferable methodology. Checkpoint-resolved, per-head analysis under fixed selection rules can expose sparse temporal organization in other factorized video diffusion transformers and, with adapted routing metrics, in joint spatio-temporal architectures.
Read the full abstract

All authors Taewoo Ha, Shafayat Mowla Anik, Dae Yeol Lee, Byeong Kil Lee, Jeeho Ryoo

Computation and Language

Distributional sentiment modeling and anomaly detection for consumer complaint assessment

Sentiment analysis is a common tool for converting unstructured text into quantitative signals in finance and risk management. Yet most applications reduce the output to a discrete polarity label or a single predictive feature, overlooking the distributional structure of sentiment intensity in consumer complaint narratives. In this paper we treat negative sentiment in consumer complaints as a bounded continuous variable and study its full distribution rather than a single label. We score each narrative with a transformer classifier, model the scores with Beta distributions, and compare the fitted distributions of meritorious and non-meritorious complaints through the Kullback Leibler divergence and the squared Hellinger distance. The fitted distributions are then linked with dollar amounts and company response outcomes to construct anomaly diagnostics that flag complaints whose textual severity is inconsistent with the recorded relief. We find that the two groups have strongly overlapping distributions, so negative sentiment intensity is not a sharp classifier of outcomes on its own; combined with monetary and categorical attributes, it isolates unusually severe complaints for operational risk monitoring. Treating sentiment analysis as continuous distributional measurement, this study links sentiment extraction, bounded response modeling, and anomaly detection in a unified framework for consumer complaint assessment.
Read the full abstract

Sound

Open-Qwen-Music: An Auditable Framework for LLM-Based Music Composition and Diffusion Rendering

We present Open-Qwen-Music, an open reconstruction of Qwen-Music and a fully specified research system for text-to-music generation that couples LLM-based semantic composition with diffusion-based acoustic rendering. The system comprises a 25 Hz single-codebook music tokenizer, a 3B-parameter autoregressive Music LLM, and a diffusion renderer producing 48 kHz stereo audio, following the cross-module interfaces reported by Qwen-Music. The strongest systems of this design remain closed, and prominent open music-generation projects release weights and inference code without their training corpora or end-to-end training implementations. This limits independent and controlled study of how information loss and prediction errors propagate from semantic representation through autoregressive planning to acoustic rendering. To our knowledge, Open-Qwen-Music is the first fully open release of an LLM-composition-plus-diffusion-rendering text-to-music system. Beyond model weights and inference code, the release includes the training datasets and provenance manifests, complete data-processing, annotation, training, inference, and evaluation pipelines, configurations, and pretrained weights for every learned module. Artifact manifests bind the identities of these artifacts across the complete workflow. Together, these artifacts establish a reproducible implementation of the modular architecture and provide an empirical basis for component-level analysis and future evaluation. We present the system as a transparent, executable research baseline and a starting point for the community, not as evidence of quality parity with Qwen-Music. Open-Qwen-Music is an ongoing effort, and we will continue to improve its generation quality, controllability, and robustness. All release artifacts are available at https://github.com/biang15343100-source/Open-Qwen-Music.
Read the full abstract

Computer Vision

PalmLeaf-VQA: A Multi-Script Visual Question Answering Benchmark for Historical Palm-Leaf Manuscript Understanding Across Diverse Regions

Historical manuscripts remain largely absent from modern vision-language benchmarks, leaving open how well multimodal large language models (MLLMs) handle culturally diverse, degraded, and non-Latin document images. We introduce PalmLeaf-VQA, a multi-script visual question answering benchmark for historical palm-leaf manuscript understanding across South and Southeast Asian traditions. PalmLeaf-VQA contains 923 curated manuscript images and 7,384 question–answer pairs from eight collection groups: Balinese, Grantha, Jathakam, Kambaramayanam, Kannada, Khmer, Sundanese, and Tamil. Unlike recognition-oriented resources, the benchmark targets manuscript-aware visual reasoning over preservation-relevant cues, including physical condition, line structure, material and coating, binding holes, margins, symbols, drawings, and localized visual artifacts. We evaluate recent proprietary and open-weight MLLMs under open-answer and constrained-answer prompting and provide fine-grained analysis across collections, question categories, and task types. The strongest evaluated model reaches only 58.00% exact-match accuracy on the held-out test split, revealing substantial limitations in current MLLMs for rare-script, degraded-layout, and preservation-oriented document understanding. PalmLeaf-VQA provides a standardized benchmark for advancing culturally grounded and layout-aware multimodal document analysis.
Read the full abstract

Computation and Language

A literature-guided descriptor-based framework for filtering composition search spaces

Scientific literature contains latent knowledge about materials behavior, but much of this knowledge is expressed through words, contexts, and recurring associations rather than explicit design principles. This raises a central question: how can large-scale scientific corpora be used for practical problems in materials discovery? Here, we present a literature-guided descriptor-based filtering framework for reducing composition search spaces. For a given performance metric, the framework selects two descriptors from a filtered vocabulary in a literature-trained word embedding model and uses the selected descriptors to construct a Pareto-based filter for candidate compositions. Across the evaluated performance metrics and composition search spaces, the framework filters out an average of 74.27% of the candidate compositions, with an average best-value error of 1.93% relative to experimental measurements. Compared with expert-chosen and random descriptors, our performance metric-dependent descriptors provide a more controlled balance between retained fraction and best-value error. These results show that literature-derived embeddings can support intuitive and reproducible filters for narrowing candidate composition spaces while preserving high-performing compositions.
Read the full abstract

Machine Learning

NeuralCert: certified computational discovery of extremal mathematical constructions

Neural networks are becoming popular in solving mathematical problems, but stochastic models do not provide mathematical exactness by themselves. This study introduces a discovery-to-certification framework in which high-dimensional variational trial functions are learned in a compact separable representation, spectrally diagnosed and pruned, and then certified exactly through multimodular evaluation. Exact certification makes the numerical proofs fully explicit and independently verifiable. This framework can be run on a standard personal computer. Across three extremal problems, we show that neural optimization can contribute to rigorous mathematics in three distinct ways: by discovering improved constructions, by exposing empirical invariants that lead to proofs, and by revealing optimization barriers whose geometry motivates new analytic or numerical representations. More broadly, these results suggest a path toward AI-assisted mathematics in which flexible computational discovery and exact certification become complementary components of a single rigorous workflow.
Read the full abstract

Computer Vision

Long-Tail Adaptive Flow Matching with Explicit Conditional Consistency Guidance for Precise Multimodal Face Synthesis

Although diffusion-based methods have substantially improved the controllability of multimodal face synthesis, their semantic alignment remains suboptimal because most existing approaches rely on implicit latent-space objectives to model the relationship between denoising variables and multimodal conditions. Such implicit modeling is often insufficient to enforce precise correspondence between synthesized faces and conditional inputs, especially under long-tailed semantic mask distributions where rare attributes receive weak optimization signals. To address these limitations, we propose EC\textsuperscript{2}Face, a multimodal face synthesis framework that improves semantic alignment through explicit semantic supervision and distribution-aware optimization. First, we introduce Explicit Conditional Consistency Guidance (ECCG), which imposes direct consistency supervision in pixel space by decoding an approximate reverse estimate of the clean latent and explicitly aligning the synthesized image with textual descriptions and semantic masks. A temporal dynamic modulation function is further designed to adapt the supervision strength according to the timestep-dependent reliability of reverse estimation. Second, we propose Long-Tail Adaptive Flow Matching (LAFM), which reweights spatial optimization signals based on semantic attribute frequency, with normalized weights to maintain numerical stability during training. Importantly, all additional modules are used only during training and introduce no extra inference overhead. Extensive experiments show that EC\textsuperscript{2}Face consistently outperforms competitive baselines in both generation quality and semantic alignment, achieving a 29.38% improvement in mask accuracy on rare attributes.
Read the full abstract

All authors Yushe Cao, Xuechao Zou, Xing Xi, Dianxi Shi, Chun Yu, Junliang Xing

Algebraic Geometry

Planar rank-one sheaves on $\mathbb{P}^3$, obstruction bundles, and divisor-supported Donaldson–Thomas series

Let $X=\mathbb{P}^3$ and let \[ \alpha_n=(0,1,-\tfrac12,\tfrac16-n)\in H^{\mathrm{even}}(X,\mathbb{Q}) \] with respect to the basis $1,H,H^2,H^3$, where $H=c_1(\mathcal{O}_X(1))$. We prove that every Gieseker semistable sheaf on $X$ with Chern character $\alpha_n$ is stable and uniquely of the form $\iota_{P*}\mathcal{I}_Z$ for a plane $P\subset X$ and a length-$n$ subscheme $Z\subset P$. Hence the coarse stable-sheaf moduli space is the relative Hilbert scheme of points on the universal plane over the dual projective space $X^{\vee}$. Using the standard perfect obstruction theory for stable sheaves on a Fano threefold, we identify the obstruction bundle with a relative Carlsson–Okounkov twisted tangent bundle and obtain a closed product formula \[ \sum_{n\ge 0} \Gamma_n q^n=\prod_{m\ge 1}(1-q^m)^{-7} \] for the natural point-inserted two-dimensional Donaldson–Thomas invariants on $\mathbb{P}^3$. We then develop the analogous divisor-supported theory for a smooth divisor $D$ in a smooth projective Fano threefold, distinguishing moving and rigid divisors. For a rigid divisor satisfying $H^1(D,\mathcal{O}_D)=0$, the divisor-supported moduli component is a Hilbert scheme of points on $D$, its obstruction bundle is the twisted tangent bundle $\mathsf{T}_D^{[n]}(N_{D/X})$, and its generating series is \[ \prod_{m\ge 1}(1-q^m)^{-(c_2(X)\cdot D+D^3)}. \] We work out the examples of the exceptional divisor in $\operatorname{Bl}_p\mathbb{P}^3$ and of a rigid section in a Fano $\mathbb{P}^1$-bundle over $\mathbb{P}^1\times\mathbb{P}^1$.
Read the full abstract

Group Theory

A conditional construction of a nonhyperlinear group and the centralizer problem

We show that a positive answer to the centralizer problem for tracial matrix ultraproducts implies the existence of a finitely generated nonhyperlinear group.

Logic in Computer Science

Gödel's and Scott's Variants of the Ontological Argument in Lean 4 and TPTP THF

The Isabelle/HOL dataset of Benzmüller and Scott's study of Gödel's ontological argument and Scott's variant (Monatshefte für Mathematik, 2025) is carried to Lean 4 and from there back to the automated provers, as a benchmark independent of either proof assistant. The port covers all thirty theories, structure and names preserved: 548 statements compare identical as parsed, every named result is proved again, and five results the original reports without replaying them are proved here. For every theorem, #print axioms gives the postulates its proof consumes: Scott's necessary existence and modal collapse need only a symmetric frame, confirming that KB suffices. The benchmark, in TPTP THF and SMT-LIB, turns the steps of an argument debated in philosophy into 294 theorems, alongside 45 statements the original refutes or leaves open, ten left open there. Five THF provers, and cvc5 on SMT-LIB, prove 227 theorems within ten seconds on one core and 232 within sixty, and none proves any of the 45. E and Leo-II solve the most, although Leo-II's calculus has been unchanged for about a decade and was only repaired and modernised here, as release 2.2. Vampire, whose later version won the higher-order division of CASC-30, solves the most in no configuration. Only E and Leo-II are measured in their own automatic mode: Zipperposition proves 101 in a single mode and 213 with its developers' portfolio, Vampire 174 without options and 209 with a higher-order schedule that its CASC mode does not select, and Leo-III 159 alone and 177 with E as partner.
Read the full abstract

2609.26806 · 14 Sep · 57 pages. Version 3 measures every prover in the setting it is used in (CASC, SystemOnTPTP, Sledgehammer), which changes several figures, and cites the companion article arXiv:2609.36279, which settles all ten statements the dataset leaves open. Ancillary files: the Lean 4 package, its typeset sources, the tools, and both renderings with every prover result

Differential Geometry

Balanced Bismut torsion-parallel manifold with constant holomorphic sectional curvature

A long-lasting conjecture in non-Kähler geometry says that any compact Hermitian manifold with constant holomorphic sectional curvature must be either Kähler or Chern flat. The conjecture is known to be true in dimension 2 but remains open in dimensions 3 or higher, except in several special cases. In this article we consider compact Hermitian manifolds that are Bismut torsion-parallel (BTP for brevity), which means that the Bismut connection has parallel torsion. For BTP manifolds that are not balanced, the conjecture was proved by Chen and Zheng. They also confirmed the conjecture for balanced BTP manifolds in dimension 3, utilizing a classification result for such threefolds by Zhao and Zheng. In a recent work, we confirmed the conjecture for balanced BTP manifolds in dimension 4 through a detailed case-by-case analysis. In this article, we establish the conjecture for balanced BTP manifolds in all dimensions. We also disprove the full-rank $B$ conjecture by constructing a compact balanced BTP fivefold with full-rank $B$ that is neither Kähler nor Chern flat; its Chern holomorphic sectional curvature is nonconstant.
Read the full abstract

Probability

Optimal complete-feedback card guessing under nonincreasing valley weights: A proof of the Diaconis-Fulman-Holmes shelf-shuffling conjecture

Let a random permutation $W\in S_n$ represent the order of $n$ distinct cards. After each guess, the player learns the actual card, and the objective is to maximize the expected total number of correct guesses. Diaconis, Fulman, and Holmes conjectured that, after a single uniform, unbiased shelf shuffle, the direction-tracking strategy is optimal: first guess the smallest card, and thereafter choose the nearest remaining card in the direction of the most recent rise or fall. We prove a more general theorem: if \[ \mathbb P(W=w)=\frac{f(v(w))}{\sum_{\sigma\in S_n}f(v(\sigma))}, \] where $v(w)$ is the number of interior valleys of the permutation, $f$ is nonnegative and nonincreasing, and the denominator is positive, then direction tracking maximizes the conditional probability of correctly guessing the next card after every history of positive probability. It therefore maximizes the expected total number of correct guesses among all complete-feedback strategies. Here an interior valley is a position $2\le i\le n-1$ satisfying $w_{i-1}>w_i<w_{i+1}$. Combined with the known shelf-shuffling probability formula, this theorem proves the conjecture for every deck size and number of shelves. The key to the proof is a swap of adjacent remaining values that preserves the revealed prefix: starting from the preferred candidate, the swap leaves the valley count unchanged or increases it by one, thereby turning monotonicity of the weights into a conditional probability ordering after every history.
Read the full abstract

Probability

Extremal expectations in card guessing with partial feedback: proofs of two conjectures of Diaconis-Graham-Spiro

We study the minimum and maximum expected scores over all adaptive strategies in sequential card guessing with yes/no feedback. A deck contains $n$ labels, each appearing $m$ times, and is shuffled uniformly. The cards are guessed one at a time, and after each guess the player is told only whether it was correct. Always guessing the same label gives exactly $m$ points. How far can the expected score depart from $m$ if all previous feedback is used either to seek or to avoid correct guesses? Diaconis, Graham, and Spiro conjectured that the minimum expectation is $m-o(m)$ as $m\to\infty$ with $n$ sufficiently large in terms of $m$, and that, for each fixed $m$, the limit superior of the maximum expectation as $n\to\infty$ is at most $(e-1)m$. Writing these extrema as $P^-_{m,n}$ and $P^+_{m,n}$, we prove \[ 0\le m-P^-_{m,n}\le3m^{3/4} \qquad(n\ge m\ge1), \] and the finite-parameter bound \[ \frac{P^+_{m,n}}m\le\sum_{k=1}^{n}\frac1{k!}<e-1 \] for all positive integers $m,n$. The first bound gives a first-order asymptotic uniform in $n\ge m$; the universal constant $e-1$ in the second cannot be reduced. Thus both conjectures follow. The lower bound combines a switching comparison for restricted permutations with martingale second-moment estimates. For the upper bound, we represent the repeated cards as a random interleaving of independent decks of distinct cards and prove that the optimal expectation is additive when the component-deck labels are revealed.
Read the full abstract

General Mathematics

Evaluations of Series Using the Beta Function

This article presents proofs of seven conjectural series identities for $\pi$ proposed by Z.-W. Sun. Our proofs rely mainly on transformations involving the Beta function. We also use symbolic integration with the software Mathematica to reduce the amount of computational work.
Read the full abstract