Computer Science & Mathematics

Long-term stability remains a critical challenge limiting the photovoltaic and optoelectronic applications of two-dimensional lead halide perovskites. In these materials, the choice of the organic spacer plays a key role not only in enhancing environmental stability, but also in determining their structural characteristics and carrier dynamics. Here, we present an environmental photo- and thermal-stability analysis of BA2PbI4 and PEA2PbI4 two-dimensional lead halide perovskites by means of confocal micro-Raman and X-ray photoelectron spectroscopy techniques. We have investigated two different organic spacers, n-butylammonium (BA) and 2-phenylethylammonium (PEA), which provide comparable room-temperature luminescence in freshly exfoliated perovskite nanoflakes and degrade similarly through the removal of organic components and fast crystal phase decomposition into PbI2 and Pb0. However, we observed that the PEA organic cation significantly improves crystal morphology and suppresses lattice defects, resulting in slower degradation and enhanced resistance to light and thermal stress under ambient conditions. Our findings indicate that structural and chemical engineering, along with surface passivation and encapsulation strategies, are essential for achieving long-term stability in two-dimensional halide perovskite devices. Nevertheless, optimizing the balance between stability and device performance remains a critical consideration.

Realizing Embodied AI for Science requires robots to translate experimental goals into reliable physical procedures, reducing hazardous and repetitive human work while generating evidence for scientific discovery. Although language-based planning and learned manipulation provide complementary capabilities, their integration requires robots to adapt subsequent operations to actual execution outcomes. For example, an unsuccessful placement changes what must be done next, even when the experimental goal remains unchanged. Existing benchmarks assess planning, manipulation, and laboratory workflows through task completion and reasoning or execution metrics; however, aggregate performance leaves the connection between outcome-dependent choices and physical execution unresolved. To address this evaluation gap, we introduce Labference, a benchmark for evaluating how observed physical outcomes guide subsequent choices and how those choices translate into robotic laboratory actions. Its evaluation protocol is designed to test outcome-dependent behavioral adaptation through controlled evidence variations, while independently measuring operation selection, physical execution, and task completion. Matched Oracle-assisted continuations from shared checkpoints then assess which forms of support improve subsequent choices and task completion. Initial weighing evaluations reveal target-selection errors in the tested VLA policies despite successful object transfers, whereas GPT-driven control achieves correct selection but encounters execution failures and substantial computation costs. Together, these findings highlight the need to connect task-appropriate reasoning with reliable, timely action. By making this connection measurable and diagnosable, Labference supports targeted improvements toward autonomous experimentation for Embodied AI for Science.

The number of paper submissions to major AI conferences has increased dramatically in recent years, substantially increasing the peer-review workload of researchers. At the same time, large language models (LLMs) are increasingly used to support peer review, both by individual reviewers and by conference organizations. While LLM capabilities have improved considerably, their evaluations may be influenced by linguistic characteristics that are unrelated to scientific content. In this work, we introduce a controlled experimental framework for studying such effects by systematically perturbing the linguistic style of research papers. Specifically, we rewrite papers to exhibit (i) linguistic features associated with native or non-native English writing, and (ii) high- or low-confidence linguistic features. To construct these interventions, we derive fine-grained rewriting instructions grounded in the linguistic literature and apply them using LLMs. Crucially, our interventions are designed to preserve the underlying scientific content. We then evaluate the resulting papers using different LLM reviewers. Our experiments expose a critical vulnerability in LLM-based peer review: reviewers systematically reward confident rhetoric independent of scientific content, raising serious concerns about the robustness of their assessments.

Geometrical flat bands provide an ideal platform for studying electronic correlations and quantum-geometric effects. Although the ferromagnetic Weyl semimetal \CoSnS is generally considered a moderately correlated system, recent spectroscopy experiments have revealed a flat-band feature near the Fermi level ($\EF$) that is absent in conventional density functional theory (DFT) calculations. Using DFT plus dynamical mean-field theory and a minimal kagome model, we show that this feature arises not from heavy-fermion-like dynamical band renormalization, but from the shifting of a $d_{x^2-y^2}$ dominated geometrical kagome flat band toward $\EF$ driven by orbital- and spin-dependent self-energies. The relocated flat band enhances the local Berry curvature near $\EF$, resulting in a strongly energy-dependent anomalous Hall conductivity and a giant anomalous Nernst effect. Our findings establish an orbital-selective correlation mechanism for positioning geometrical flat bands at $\EF$, and propose anomalous thermoelectric transport as a sensitive probe of flat band induced topology.

Navigational search aims to find a single target page among hundreds of millions of web pages. Click tables work well for queries seen before but cannot generalise to new formulations, while embedding retrievers generalise across content but often rank navigational targets poorly. We introduce Octen-WebGR, a generative retrieval framework that stores web pages in the parameters of a language model: given a query, the model generates a two-token semantic identifier constrained to a table of real URLs. We develop Octen-WebGR-4B, which indexes 201 million web pages, and evaluate it on real navigational traffic from the month after training. Generative retrieval is most effective on new ways of asking for familiar pages: for unseen queries targeting pages with substantial interaction history, it ranks the correct page first 30.9% of the time, against 17.2% for a dense retriever trained on the same clicks (21.4% with a click prior). Embedding retrieval keeps an advantage on rarely visited pages, so the two approaches generalise on complementary sides, the query and the page. Over all navigational traffic, Octen-WebGR reaches 59.6% top-1 accuracy, against 46.5% for a click-enhanced dense retriever and up to 37.5% for frontier LLMs answering from memory, which cost at least 40 times as much per query. The 4B model runs on a single GPU with 10 GB of memory, compared with 412 GB for exact dense retrieval. Scaling experiments with up to 32 million URLs show that, when every query-page pair is seen equally often, the model size needed grows sublinearly with index size, and that retrieval is constrained more by page exposure than by model capacity. We release Octen-WebGR-4B and Octen-WebGR-0.6B with their identifier tables and inference code.

Idioms have long been a challenge for machine translation, and while large language models (LLMs) have proven increasingly adept at handling them, whether quality estimation (QE) systems have kept pace and can reliably score idiomatic renderings of these expressions above faulty literal ones remains unclear. We introduce the first parallel dataset pairing idiom-bearing English sentences with both literal and idiomatic translations across nine target languages, augmented into multi-sentence segments and a graded literality continuum. Using this dataset, we benchmark eleven conventional QE metrics from the COMET and MetricX families and ten LLMs from the Qwen-3, Gemma-4 and OLMo-3 families used as GEMBA-style judges. We find that most systems, especially older metrics and smaller models, display a marked literality bias, rating literal translations of idioms above idiomatic ones. Higher parameter counts and newer training sets appear to reduce this bias to an extent; nevertheless, of all the benchmarked systems, only MetricX-24-Hybrid (XL, XXL) and Gemma-4-31B consistently rate idiomatic translations of idioms above literal ones, and only Gemma-4-31B reliably ranks intermediate degrees of idiomaticity and literality. Our findings caution against deploying standard QE metrics in domains rich in figurative language, where higher scores might be rewarding undesirable literalism.

Conversational bandits improve recommendations by actively querying users about key-terms to elicit their item preferences. However, existing methods typically assume homogeneous feedback for items and key-terms, using either absolute rewards or pairwise comparisons. In practice, item feedback often takes the form of clicks or ratings, while comparisons are more natural for key-terms. To address this gap, we study conversational generalized linear bandits with hybrid feedback, combining absolute item rewards and relative key-term comparisons. We propose Conversational UCB with Hybrid Feedback (HyConUCB), which constructs unified confidence bounds from heterogeneous feedback and adaptively selects key-term comparisons by balancing information gain and directional coverage. Through a refined regret analysis that jointly exploits both feedback sources, we establish a regret bound that improves upon reward-only and adapted dueling-only benchmarks. Experiments on synthetic and real-world datasets validate the effectiveness of our algorithm.

Natural-language-to-domain-specific-language (NL-to-DSL) semantic parsing systems increasingly translate informal user descriptions into structured artifacts such as Structured Query Language (SQL) queries, Application Programming Interface (API) calls, workflows, formal specifications, and ontology-grounded domain-specific language (DSL) statements. This paper focuses on ontology-grounded Assertion Box (ABox) DSL generation: the translation of natural language domain descriptions into assertion-level DSL statements grounded in a pre-existing ontology vocabulary. In this setting, syntactic validity is often easier to achieve than semantic adequacy. Failures are frequently attributed to weak generator models, but recurring sources of difficulty are also structural: insufficient input, weak domain grounding, output template and vocabulary gaps, weak validation, and non-local repair. We propose a role-level difficulty profile for diagnosing difficulty in ontology-grounded NL-to-DSL semantic parsing at the workflow level. The profile organizes diagnostic variables across five recurring workflow roles: Input, Generator, Output, Validator, and Repairer. The seven variables are input sufficiency, context availability, grounding complexity, template fit, vocabulary coverage, validation strength, and error observability. Practitioners construct the difficulty profile through qualitative judgment over available evidence. This experience-informed, non-exhaustive difficulty profile is intended to support workflow diagnosis and guide redesign or human-in-the-loop intervention.

Lean verifies that a formal proof establishes its stated proposition under declared assumptions, but does not check whether that proposition faithfully expresses the mathematical claim in a manuscript. Assessing this correspondence requires interpreting the main theorem through its project dependencies, which may be scattered throughout a large development. We introduce MechAlign, a framework for project-level semantic alignment review that combines formal dependency extraction with LLM-based assessment. Its extraction tool, LeanTrace, constructs a statement dependency graph from elaborated declarations, retaining definition bodies and structure constraints while excluding proof bodies and stopping at an explicit library boundary. The reviewer uses this context to investigate alignment and report discrepancies with localized evidence. We conduct a preliminary review of 201 main results from 173 OpenAI Math projects, using supplied informal descriptions and available paper context. A hyperbolic-group case illustrates how dependency inspection localizes a missing dimensional constraint relative to the manuscript, consistent with the upstream declaration of formalization scope. We also construct LeanReviewBench, comprising 501 fixed claim-statement-context triples from OpenAI Math, Fermat's Last Theorem, and Kakeya, with description provenance recorded. The resource supports controlled comparisons of more project-level semantic alignment evaluations.

Optical coherence tomography (OCT) is a non-invasive imaging modality used for visualization of retinal tissue. However, OCT volumes are affected by speckle noise, which degrades image contrast and obscures fine structural details. Furthermore, modern clinical practice requires large field-of-view, high-resolution OCT volumes, whose storage and transmission are demanding for archiving and teleophthalmology applications. To address aforementioned challenges, we propose a low-rank multi-scale entanglement renormalization ansatz (MERA) tensor network approximation of 3D OCT image volume, where the image volume is represented as contractions of a top core factor encoding global features and multiple interacting low-rank orthogonal or semi-orthogonal factors modeling inter- and intra-B scan correlations. This representation jointly enables compression, denoising, and feature extraction. Unlike deep learning approaches, the proposed approach is training-free and processes each OCT volume independently. We evaluate the proposed framework on OCT image volumes acquired by two different scanners and demonstrate improved visual quality and reduced retinal layer segmentation errors in comparison with several competing methods. We also extract MERA features from two clinical OCT volume datasets. They are subsequently classified using a Grassmann-manifold nearest-subspace classifier. With the subspace dimension as its only hyperparameter, the classifier is theoretically shown to be robust in small-sample clinical scenario. We demonstrate competitive accuracy of classification of retinal diseases in these small-sample settings, where deep learning methods are often limited by the scarcity of labeled data. The proposed structure-aware model-driven representation learning for high-dimensional data under limited supervision provides good generalization performance while maintaining high interpretability.

A transcriber listening through one microphone also hears everyone near it. We train causal voice isolators: masking transformers of 140M, 243M and 374M parameters, 20 ms of latency, enrolling the caller from the call itself. Short evaluation windows hide two failures that real calls expose: forgetting the caller when another voice talks alone for a while, and damaging a caller who is alone. Long calls with turns, memory beyond the attention window, real headset crosstalk and drifting caller levels address both, but each trades one against the other. WavLM-Large's own weights, made causal and trained with the model, move the trade: M with them (555M parameters) lowers the error on our development sets from 11.44% to 10.55%; a student trained from scratch to imitate WavLM does not, and w2v-BERT 2.0 and Whisper's encoder, pre-trained on far more audio, help less or not at all. A lone caller loses most at the start of a recording: no training call opened with someone already talking. Training on windows cut from longer calls lowers the error on clean development calls from 3.92% to 2.96%; this is our release. Live, the release makes 16.13% word errors on 60 s AMI windows, against 23.66% for VoxLock, the strongest open isolator, and 10.19% on its own AMI windows, against 11.71%; it is level on Krisp's benchmark (5.74% against 5.63%), and a caller who is alone still loses more words than with VoxLock (5.58% against 2.92% on clean test calls). Distilled into 22.1M parameters without WavLM, it runs live on one CPU thread inside a LiveKit agent, at 33% of a core per call in 20 ms frames, with 10.66% error on our development sets (the release 10.49%). We open-source the release and this CPU model, with a LiveKit plugin, at https://huggingface.co/collections/Scicom-intl/voice-isolator.

Thermal processing of nickel (Ni) layers on germanium-tin (GeSn) can strongly modify the Ni film and the Ni/GeSn interface within a restricted thermal budget; however, the influence of GeSn crystalline quality on Ni-Ge-Sn interfacial evolution and the magnetic response of the Ni layer remains incompletely understood. Here, Ni (15~nm)/Ge$_{1-x}$Sn$_x$/Ge/Gd$_2$O$_3$/Si(111) heterostructures ($x = 0.08$ and $0.10$) were investigated following rapid thermal annealing at 573 and 973~K for 10~s using X-ray diffraction, pole-figure analysis, X-ray photoelectron spectroscopy depth profiling, and temperature- and field-dependent magnetization measurements. Annealing at 573~K preserves the crystallographic texture of Ge$_{0.92}$Sn$_{0.08}$, whereas Ge$_{0.90}$Sn$_{0.10}$ exhibits microtwin formation and reduced structural integrity. Annealing at 973~K, employed as a severe upper-bound thermal reference, causes extensive structural degradation in both heterostructures, with substantially greater deterioration in the higher-Sn sample. XPS depth profiling reveals progressive Ni-Ge-Sn intermixing accompanied by pronounced Sn surface segregation with increasing annealing temperature. Magnetization measurements show coexisting ferromagnetic and superparamagnetic contributions in the as-deposited and 573~K samples, whereas annealing at 973~K strongly suppresses the ferromagnetic response, producing highly unsaturated $M(H)$ loops together with reduced saturation magnetization and coercivity. These magnetic changes are consistent with extensive interfacial alloying and dilution of the metallic Ni layer. This study establishes a direct correlation between GeSn crystalline quality, interfacial diffusion, and magnetic behavior.

We consider decentralized stochastic nonconvex optimization over directed networks. Push--Pull provides an efficient decentralized optimization framework for such networks, but in the stochastic setting, gradient variance can slow its convergence and complicate the analysis of linear speedup. Motivated by this observation, we incorporate recursive momentum into stochastic Push--Pull for variance reduction, yielding the accelerated stochastic Push--Pull (ASPP) algorithm. Under mean-squared smoothness, we prove that ASPP improves the convergence rate of stochastic Push--Pull and achieves asymptotic linear speedup with respect to the number of agents. To further reduce communication, we develop LASPP, a local-update variant of ASPP that allows each agent to perform multiple local updates between communication rounds. We show that LASPP achieves an additional asymptotic linear speedup with respect to the number of local steps. With a suitable number of local steps, LASPP reduces communication complexity while maintaining the same stochastic gradient complexity as ASPP. Numerical experiments on real datasets demonstrate the effectiveness of the proposed algorithms.

Embedding-based verification often calibrates a false-accept threshold from all impostor pairs among $m$ identities, although these pairs share identities. We characterize a worst-case limit: when hub identities score highly against a fraction $ρ$ of the population, a calibration-only threshold with marginal false-accept rate (FAR) at most $α$ under every identity-exchangeable law reaches the hub score on hub-free calibration sets with probability at most $(m+2)α/(2ρ)$. Uniformly accepting near that score at levels well below $ρ/m$ therefore requires trial information. We propose COVE, which ranks the trial pair among all $N=\binom{m+2}{2}$ pairs formed by augmenting the calibration units with the probe and reference. Under identity exchangeability, its marginal FAR is at most $\lfloorαN\rfloor/N$, with first possible acceptance at $1/N=O(1/m^2)$. The rule also retains this guarantee with a fixed symmetric cohort-normalized score using an independent cohort. On 6 benchmarks and 8 frozen encoders, realised FAR was consistent with the exact bound. Where both valid calibration-only baselines accepted nothing, COVE retained a median $100\%$ of the sampled oracle's true-accept rate over reliable comparisons. In contrast, an empirical quantile over dependent pairs reached 2.87$\,α$ FAR, and a binomial upper confidence bound over pairs missed its nominal 95\% conditional-FAR target in up to 55\% of calibration draws. Symmetric CSLS raised COVE's true-accept rate by up to 4.3-fold in the tested settings. The code is available in https://github.com/salimkhazem/Cove.git

Self-supervised pretraining has produced capable time-series foundation models, but its value for Prognostics and Health Management (PHM) has not been measured under leakage-controlled conditions. We measure it on a corpus of 47 public PHM datasets, harmonized to a single tensor contract and partitioned so that 11 datasets are withheld from pretraining in full. Transfer turns out to be task-dependent: a compact channel-independent Transformer encoder (801,808 parameters) pretrained with masked patch prediction raises fault-classification macro-F1 from 0.350 to 0.812 on CWRU and from 0.569 to 0.951 on HSG18, yet on CMAPSS remaining-useful-life estimation no pretrained checkpoint reliably improves upon a from-scratch baseline. We pretrain under two regimes---centralized, and simulated cross-silo federated with ten clients grouped by physical domain---holding architecture, objective, source data and optimization budget fixed. Two further findings are negative. At matched budget the centralized encoder leads by 15.5--20.2 macro-F1 points in four of four comparisons, although the comparison is confounded by a learning-rate schedule that favours the centralized arm and rests on a single pretraining run per regime; on RUL the two regimes are indistinguishable. And the intrinsic SSL loss is not a valid selection criterion: the centralized encoder more than halves the federated reconstruction loss without a corresponding gap on RUL, and a server-momentum sweep moves downstream validation macro-F1 by up to 0.29 while the SSL loss varies by 0.021. Transfer also hinges on the backbone learning rate: the same checkpoint yields 0.169 or 0.829 macro-F1 on CWRU depending on a single order of magnitude. We release the corpus pipeline, the protocol and the code.