Papers

We report an exploratory, single-seed ablation of TALH (Adaptive Latent Hybrid), a decoder-only language model with parallel Multi-head Latent Attention (MLA) and a custom recurrent state-space (SSM) branch. Five variants, spanning 117–217M estimated active parameters per token, are trained from scratch on a FineWeb sample for the same number of optimisation steps and tokens. In this specific setup, removing the SSM branch gives the largest degradation in validation perplexity (MLA-only PPL 315), whereas removing MLA has a much smaller effect (SSM-only PPL 239). A dense-FFN hybrid obtains PPL 231, compared with 240 for the tested top-2 ternary-MoE hybrid, while using 3.87 GB less peak training memory. We also preserve a preliminary Apple M3 timing observation: among the five unoptimised implementations, MLA-only has the flattest measured time-to-first-token curve from 512 to 2,048 prompt tokens, although the dense Transformer is much faster in absolute terms. Because the runs are single-seed, parameter counts are unmatched, the evaluation stream may overlap the training source, and raw repeated timing records are unavailable, these results support implementation-specific hypotheses rather than general conclusions about MLA, SSMs, or mixture-of-experts models.

We prove, to our knowledge, the first unconditional cumulative closure estimates for the Arnold–Mukherjee–Pouly (AMP) refinement algorithm and the original face-marked tetrahedral algorithm of Bänsch on arbitrary conforming initial tetrahedral meshes. Let $T_0,\ldots,T_L$ be an adaptive mesh sequence generated by either algorithm, with $M_\ell\subseteq T_\ell$ denoting the marking set at step $\ell$. Then $$\#T_L-\#T_0\leq C_{\mathrm{clos}}(T_0)\sum_{\ell=0}^{L-1}\#M_\ell.$$ The proof is intrinsic to the physical three-dimensional mesh and requires neither an initial compatibility condition nor a higher-dimensional embedding. It organizes conformity refinements into a causal forest and combines a uniform horizontal-propagation estimate with a weighted packing argument to obtain an explicit closure constant. For the original Bänsch algorithm, every history-dependent resolution of the initial two-edge ambiguity is represented by one of finitely many AMP histories. The estimate therefore holds uniformly for arbitrary deterministic or nondeterministic choices. This resolves a long-standing complexity question for the Bänsch–AMP family.

This paper introduces an innovative and bias-free variant of a three-player auction bridge, specifically designed to address structural limitations found in traditional formats. In a standard auction bridge, the partner of the winning bidder often plays a passive role, possessing minimal influence over the outcome once the bidding phase concludes. To rectify this, our proposed three-player model redistributes strategic agency, emphasising a framework where No Trump strategies statistically outperform specific trump card selections. The core of this research is a rule-based dual-algorithm framework. First, we implement an iterative bidding algorithm that allows players to calibrate their bids dynamically based on evolving hand strengths. Second, we introduce a predictive iterative algorithm designed to calculate the expected number of tricks a player will secure before the lead card is played. Unlike traditional approaches that rely on backtracking or high-complexity dynamic programming, our iterative method significantly reduces time complexity. This efficiency is crucial for real-time gameplay, enabling players to perform complex probability assessments instantly without computational lag. To validate the model, a simulation was conducted on a dataset of 1,00,000 game instances. This large-scale analysis allowed for a precise comparison between initial bidding, outcomes, and overall bidding accuracy. The empirical results demonstrate that our algorithm not only enhances the balance and fairness of the auction bridge but also exhibits high reliability in predictive accuracy. Beyond the realm of card games. By modelling uncertainty and strategic competition, the algorithm offers valuable insights into risk management and decision-making in volatile market environments.

We present a numerical formulation for the resolution of finite systems of interacting edge dislocations and wedge disclinations. The approach is based on a finite element discretization of a fourth-order elliptic boundary value problem arising from the mechanical equilibrium equations of plane-strain linear elasticity in the presence of kinematic incompatibilities. The numerical implementation follows a continuous interior penalty discontinuous Galerkin framework and relies on the solution of a finite set of local cell problems. Our method is validated against analytical benchmarks and used to explore several non-trivial dislocation-disclination configurations, illustrating how translational and rotational incompatibilities shape the stress state of the body.

We present a modular and multi-level framework for the differential diagnosis of malignant melanoma. Our framework integrates contextual information and evidence at the lesion, patient, and population levels, enabling decision-making at each level. We introduce an anatomic-site aware masked transformer, which effectively models the patient context by considering all lesions in a patient, which can be variable in count, and their site of incidence. Additionally, we incorporate patient metadata via learnable demographics embeddings to capture population statistics. Through extensive experiments, we explore the influence of specific information on the decision-making process and examine the tradeoff in metrics when considering different types of information. Validation results using the SIIM-ISIC 2020 dataset indicate including the lesion context with location and metadata improves specificity by 17.15% and 7.14%, respectively, while enhancing balanced accuracy. The code is available at https://github.com/narenakash/meldd.

Range query is a fundamental task in geospatial data search and many other downstream applications. Classic range queries often rely on tree-based spatial indexes, of which the query speed depends on the number of indexed points $k$ within the queried range. For instance, a classical B+ tree answers a range query in O(log N+k). For a long time, this speed has long been considered asymptotically optimal in classic database systems, until the recent emergence of quantum computing, where a quantum B+ tree may requires only O(log_B N). This paper presents Quantum Range Query (QRQ) via a hybrid quantum-classic algorithm to return the range query results in quantum superpositions. In this context, QRQ is designed to accelerate classic range query on spatial-temporal-semantic trajectory geodata using quantum algorithms. Specifically, QRQ develops quantum variants of R-tree, TB-tree and KD-tree, where the physical slots of a node, including unused padding slots, are treated as an array that a quantum random-access memory (QRAM) can read in superpositions. Evaluations on three common trajectory datasets, namely GeoLife, T-Drive, and GDP Drifter, and 10,000 queries per setting, the QRQ speedup at 1% target selectivity ranges from 2.03 times) to 64.70 times. More importantly, QRQ shows a great potential in optimizing existing trajectory range query, so it becomes the shared primitive on which human mobility analysis, and later searches specified by large language models (LLMs) and geo-foundation models (GeoFMs), can rest.

National cybersecurity and digital-governance capacity frameworks, most prominently the Cybersecurity Capacity Maturity Model for Nations (CMM), shape hundreds of millions of dollars in donor-funded governance investments across the Global South. Yet a persistent, under-theorised gap remains between state-level institutional maturity and the governance actually experienced by end-users. We term this the last-mile governance gap: the structural space between what a maturity assessment can see - laws, agencies, standards, awareness campaigns - and what a household living under that formal architecture can access, understand, or enforce. Drawing on a dual vantage point combining CMM assessment experience with peer-reviewed empirical fieldwork on smart-home privacy governance in Jordan, corroborated against studies from Kenya and China, we identify four mechanisms by which national capacity fails to reach the household: legibility, intra-household power distribution, accessibility of redress, and infrastructure-affordability constraints. We map these mechanisms explicitly onto the CMM's five dimensions, propose four concrete, low-cost last-mile indicators pilotable within existing CMM deployments, and outline a staged adoption roadmap with responses to anticipated objections. With AI-enabled devices entering homes across the Global South faster than institutional capacity can adapt, the stakes are rising: this dynamic risks converting a measurable maturity gap into an invisible one. We conclude with actionable implications for the GCSCC, the ITU, the World Bank, and other institutions relying on maturity scores to prioritize investment.

We establish a uniform inf-sup stability estimate for the Scott-Vogelius finite element spaces on uniform Freudenthal tetrahedralizations of the unit cube for polynomial degrees k >= 4. This result completely settles the first conjecture of Farrell, Mitchell, and Scott for the critical degrees k = 4 and k = 5, complementing the known stability range for higher polynomial degrees. The main mathematical difficulties stem from the complex topological compatibility required at the singular vertices and the corresponding mean-value constraints across adjacent elements. We tackle these challenges by developing a unified barycentric skeleton-bubble calculus that explicitly constructs vertex jets, edge modes, and face transfers to globally route element means. The accompanying exact computations independently verify these finite-dimensional identities and provide reproducibility data.

While multimodal large language models (MLLMs) have advanced video understanding, they remain highly prone to hallucinations in dynamic scenes. We argue this stems from a failure in spatio-temporal monitoring, the ability to persistently track object identities, states, and relations over time. Existing benchmarks obscure this deficit by relying on single final-answer evaluations for queries that can often be resolved via local visual cues or statistical priors. To rigorously diagnose this, we introduce STRAND, a benchmark of human-verified object-centric facts that evaluates intermediate reasoning by decomposing queries into sub-questions, distinguishing genuine temporal understanding from coincidental correctness. Crucially, we score models with Faithful Accuracy, an unconditional joint metric that credits a prediction only when the target answer and every prerequisite sub-question are correct, so that a model cannot inflate its score by being selectively consistent on the small subset of targets it happens to answer correctly. To address failure modes exposed by STRAND, we further propose an object-centric framework that explicitly constructs and reasons over structured object trajectories via chunk-wise state extraction and temporal aggregation. Extensive experiments, including backbone-, frame-, call-, and token-matched comparisons against both end-to-end MLLMs and modular video harnesses, demonstrate that our object-centric framework significantly reduces hallucinated answers and improves spatio-temporal reasoning consistency over state-of-the-art MLLMs. The code, model, and data have been made available at nguyentthong.github.io/strand.

Enforcing finite-time persistence specifications in Signal Temporal Logic typically requires formulating Control Barrier Functions within an extended state-time space. This introduces strict mathematical assumptions on global clock synchronization, rendering multi-agent execution brittle under asynchronous network delays, clock skew, and non-monotonic jitter. In this paper, we prove that for bounded-time safety tasks, explicit time dependency can be mathematically eliminated for general continuous-time nonlinear systems. By leveraging the reciprocal integral of worst-case system dissipation, we systematically compile temporal specifications into static, lower-dimensional spatial invariants. We formally prove that this time-invariant realization is strictly sound, removes the need for heuristic time-varying envelope tuning, and guarantees robust forward invariance even under severe network desynchronization.

In medical visual question answering (VQA), hallucinations of vision-language models (VLMs) may lead to confident but incorrect responses, raising the risk of diagnostic errors. Existing hallucination detection methods uniformly estimate the reliability of VLM outputs from response consistency or visual evidence. However, such uniform verification across questions ignores question-specific characteristics, resulting in missed overconfident errors and false alarms from over-verification. We present PROVE (Proof-guided Regime-aware Operator Verification), a black-box detector that adapts verification strategy to the evidential structure of each question. PROVE classifies questions into three verification regimes based on what kind of visual proof they demand, activates a regime-specific subset of five complementary operators, and adjusts operator importance per question through a lightweight calibration layer conditioned on deterministic question-answer features. PROVE uses question-specific evidence to reweight operators and produce a calibrated risk score. Evaluated on 8048 test samples across three medical VQA benchmarks and four frontier VLMs, PROVE achieves 0.821 AUROC, outperforming the strongest baseline by +0.159, with consistent gains across all models and benchmarks.

Named Data Networking (NDN) replaces address-based forwarding with longest-prefix lookup over variable-length hierarchical names, creating a difficult fit with bounded hardware fast paths. We formulate this problem as certified fast-path coverage. An offline mapper selects a prefix-closed set of parser-visible trie nodes and a separate set of action-equivalence certificates under a shared budget. A certificate relates a selected shallow FIB prefix to a deeper true-LPM class with the same forwarding action; applying the shortcut requires a sound class guard, while uncertain traffic falls back to an exact path. We present BPC-Greedy, a certificate-aware placement heuristic, and BPC-Portfolio, a multi-strategy wrapper spanning exact-heavy, repair-based, and shallow-shortcut constructions. The formulation exposes a tree-closure knapsack structure for prefix placement, exact top-k certificate selection for a fixed selected set under uniform certificate cost, and a separate bank-feasibility layer. On a synthetic parser-depth sweep and a 50k replay of public NDN testbed names under a reproducible synthetic forwarding-action snapshot, modeled certified coverage increases from 46.48% for the best exact-prefix-only bank-aware baseline to 64.06% at total budget M=48. These results characterize placement opportunity under the stated model: the evaluator uses the known true LPM as an oracle-style guard, certificate-table bank traffic is not modeled, and no claim is made about synthesized latency, area, or throughput.

Solving ill-conditioned linear systems needs LU decomposition in precisions beyond binary64. The standard approaches – GMP/MPFR, or multi-component arithmetic such as double-double – leave every scalar multiply-add inside the O(n^3) update in multiple precision, and so cannot exploit the low-precision matrix engines that dominate current hardware. We build a blocked, partially pivoted LU decomposition on Ozaki scheme II, which replaces the multiple-precision GEMM by exact integer modular products followed by an explicit CRT reconstruction, and implement it for multi-component (DD/TD/QD) and arbitrary precision, on CPUs and GPUs. Two contributions make this practical: a direct conversion between the non-overlapping expansion format and the internal fixed-point representation, which removes the MPFR round trip and is worth a factor of 2.4–4.7; and new FP16, FP8 and binary64 GPU back-ends, the FP8 one using a balanced base-17 two-digit encoding that keeps the full 362.8-bit CRT capacity of INT8. On an Arm/GB10 and an x86/H100 the fastest back-end changes with the machine: INT8 wins on GB10, whereas on H100 binary64 is fastest for QD, beating both the native implementation and INT8. The essential point of Ozaki scheme II is thus not to use a low-precision engine but to choose the format maximising bits-per-modulus times engine throughput. For the Lotkin matrix at p ~ 1.2 log2 cond(A) we reach relative errors of 10^-645 at n=2048, up to 2.84x faster than a fully OpenMP-parallel multiple-precision LU. We also show by measurement that the O(p) advantage of scheme II over scheme I applies only to the GEMM term: modular reduction and CRT grow as O(p^2) and dominate the runtime at the matrix sizes considered here. The implementation is released as open source.

Agentic AI is sold on a simple promise: the system keeps acting when the user stops watching. That promise creates an accountability inversion. As stepwise supervision recedes, verification does not disappear; it moves into the runtime infrastructure that defines authority, records action, interrupts execution, checks outcomes, and supports repair. We call this the reduced-supervision paradox. Using a 63-artifact audit, we examine its public visibility across 46 research papers and 17 engineering, documentation, security, and governance sources. We find that agents' action surfaces are far easier to reconstruct than the mechanisms needed to answer for their actions. Tool mediation and monitoring traces were clearly visible in 40 and 37 artifacts, whereas checkpoint placement was clearly visible in 6, validator independence in 4, recovery in 2, and contestability in 1. Three action paths show why this imbalance matters. A repository path can preserve rich diffs after a consequential change. A browser path can cross organizational boundaries faster than permissions travel. A service path can follow policy while leaving affected people without recourse. We argue that observability can become a substitute for accountability when it shifts verification onto users after meaningful intervention is no longer possible. Our action-path diagnostic instead asks whether a delegated action remains connected to authority, evidence, interruption, independent judgment, recovery, and challenge. The claim is deliberately bounded to public visibility; it does not establish the prevalence or effectiveness of undisclosed controls. We contribute an action-level account that recasts the harness from a technical wrapper into accountability infrastructure.

AI resurrection is often framed as a question of consent: did the represented person authorize being made to speak? That question matters, but it freezes authority at the moment of creation. A representation can later change models, pass to relatives, depend on a provider, incorporate records shared with others, or circulate far beyond its intended audience. We argue that the central governance problem is therefore not whether authorization exists once, but whether it remains legible as the representation moves. Across a public-record audit of 93 systems, creation was far easier to inspect than the conditions for speaking, contesting, preserving, or leaving: consent or authority information was thin in 82 systems, objection or redress in 82, and deletion or export in 77. The differences among systems reveal why these gaps cannot be reduced to one transparency score. Human afterlives show consent becoming incomplete over time. Companion-animal afterlives begin where subject consent is unavailable and shared care must allocate authority. Adjacent persona and mimetic systems show how voices, likenesses, and personalities can travel into later afterlife uses. Four public cases follow the same movement from premortem participation, through intimate postmortem creation, to third-party circulation and family contestation. From this evidence we develop relational authority: authorization is distributed across people, records, providers, and audiences, and must remain traceable as those relations change. This reframes AI resurrection from a product authorized once into an accountability chain linking creation authority, source boundaries, circulation, contestation, and exit. The study measures what users and affected parties can inspect publicly; private implementation and lived outcomes remain open empirical questions.

Code understandability is a critical aspect of software quality. Prior research has largely focused on this attribute from a human-centric or code-centric perspective, while it should be viewed as a relational property arising from the interaction between a reader and the code. With the increasing adoption of large language models in software engineering, we posit that the notion of "reader" should be generalized to encompass both humans and models. Building on this relational perspective, we extend the concept of code understandability to distinguish between human and model code understandability, aiming to investigate the behavioral alignment between the two. To this end, we use a dataset from a prior study containing human-rated judgments of understandability across diverse participant groups, and evaluate multiple open and closed-source LLMs. To operationalize such behavioral alignment, we introduce four behavioral proxies of model-code understandability (BPMU): P0, based on program comprehension-focused question answering; and P1–P3, based on program intent summarization (derived from our proposed semantic self-consistency). Our findings show that LLMs exhibit stronger behavioral alignment with general human code understandability than previously used shallow machine learning baselines. Stratified analyses further reveal that model code understandability aligns most closely with professional developers, but significantly less with other student groups. Role-conditioned prompting does not improve the alignment for the latter, suggesting that current LLMs cannot accurately mimic the perspectives of human readers with varying expertise. Finally, we demonstrate that semantic self-consistency is a reliable and extensible measure to be used as a behavioral proxy for quantifying model code understandability, with broad implications in both software engineering research and practice.

The Schrodinger equation in one spatial dimension admits a small set of exactly solvable potentials that serve as natural proving grounds for any new eigenvalue solver. We formulate Physics-Informed Neural Networks (PINNs) and Physics-Informed Quantum Neural Networks (PIQNNs) for the time-independent Schrodinger equation and apply them to three of these benchmarks: the harmonic oscillator, the infinite square well, and the finite square well. In each case a composite loss encodes the differential-equation residual, the normalization condition, the boundary behavior, and the orthogonality between eigenstates, so that the trial wave function is driven toward a genuine eigenfunction without supervised data. The eigenvalues and wave functions returned by both methods are compared against the exact spectra and against three classical references: the matrix Numerov method, the finite difference method, and the shooting method. For the smooth oscillator the two neural solvers reproduce the lowest four eigenvalues to parts per million, while for the square wells they recover the analytic levels with comparable fidelity even where the potential is discontinuous. The quantum circuit, built as a layered angle-embedding ansatz with strongly entangling blocks, converges more reliably than its classical counterpart on the higher excited states, where the loss landscape of the classical network becomes harder to navigate.

Large language models (LLMs) are increasingly integrated into AI-assisted hiring pipelines, including automated resume generation and screening. Under the EU AI Act, the hiring domain is classified as high-risk, making fairness and transparency critical requirements. Existing work has primarily focused on explicit hiring decisions, while less attention has been paid to whether generated resumes themselves encode recoverable demographic information. In this work, we conduct a two-stage audit of demographic leakage in German-language LLM-generated resumes. First, we use ChatGPT (GPT-4o-mini), Gemini 2.5 Flash-Lite, and multiple scales of the open-weight Qwen 3 model family (4B, 8B, and 14B) to generate resumes from real anonymized job-matching profiles, systematically varying gender- and ethnicity-associated names while holding qualifications constant. Second, we simulate a downstream resume screening scenario, where the generated resumes are first anonymized and gender-neutralized, before demographic leakage classifiers are trained on the resulting texts. We find that, despite these interventions, classifiers reliably distinguish between resumes generated with male and female names. This leakage is not driven by overtly gendered wording, but by subtle differences in the usage of semantically equivalent, formally gender-neutral terms in German. In contrast, ethnicity-related leakage remains comparatively weak across models. Our findings demonstrate that apparently neutral resume generation can still preserve highly predictive demographic signals, raising concerns about anonymization-based fairness interventions in multilingual AI hiring pipelines.

Spatial transcriptomics (ST) provides spatially resolved gene expression profiling but remains expensive, motivating the prediction of ST from histology images. Generative models have emerged as a mainstream paradigm for ST prediction due to their ability to model the conditional distribution of gene expression and capture its inherent stochasticity. However, these methods typically treat genes as independent prediction targets and overlook the intrinsic gene-gene interactions in biological systems, which limits their ability to preserve biologically meaningful co-expression patterns. We argue that gene-gene interactions, which reflect shared pathways and regulatory mechanisms, are essential for generating numerically accurate and biologically coherent ST profiles. In this paper, we propose CorrFlow, a correlation-guided flow matching framework for histology-to-ST prediction that explicitly models gene-gene dependencies through two complementary mechanisms. First, we introduce an annealed masked flow matching strategy, where subsets of genes are progressively masked following a timestep-dependent annealing schedule, encouraging the model to infer masked genes conditioned on the remaining genes and promoting joint conditional modeling beyond per-gene marginal estimation. Second, we devise a gene graph-regularized optimization scheme that integrates prior knowledge from the STRING database and data-driven co-expression estimated by WGCNA to construct a gene affinity graph, which enforces both local consistency and global smoothness in the predicted expression. Extensive experiments across 12 datasets show that CorrFlow achieves the best average PCC and HPCC among evaluated methods, leading to more biologically coherent ST predictions.

Physics-informed neural networks (PINNs) provide a mesh-free approach for solving differential equations, but their performance can depend strongly on loss weighting, collocation placement, and optimization strategy. This study develops an adaptive PINN framework for the Blasius boundary-layer equation using gradient-norm-based adaptive loss weighting, nonuniform and residual-based collocation, and sequential Adam–L-BFGS optimization. In the representative run using the architecture $[1,100,100,1]$, the model predicts $f''(0)=0.3320762918$, compared with the high-accuracy benchmark $0.332057336215$, giving an absolute error of $1.896\times10^{-5}$. The final weighted loss is $6.789\times10^{-8}$, and the predicted stream-function, velocity, and shear profiles agree closely with an independent numerical boundary-value solution. A separate full-training architecture study shows that the two-hidden-layer model achieves the smallest wall-shear error among the four tested architectures, $1.629\times10^{-6}$, whereas the deepest network attains the smallest weighted objective but a substantially larger wall-shear error. Compared with the previously reported PINN value $f''(0)=0.33165$, the representative run reduces the wall-shear error by approximately a factor of $21.5$. The results show that the combined adaptive training framework can achieve high accuracy for the Blasius problem and that weighted loss alone is insufficient for identifying the most physically accurate PINN. Because the adaptive components are applied jointly, their individual contributions cannot be isolated from the present results and would require a controlled ablation study for separate assessment.

Drug-target relation prediction supports candidate screening, drug repositioning, and mechanism analysis. Existing models often use incomplete drug or protein representations, model cross-modal interactions shallowly, or train affinity regression and interaction classification separately, although these tasks describe closely related views of the same drug-target pair. This paper presents DPTM-DT, a dual-pretrained Transformer framework for multitask drug-target prediction. DPTM-DT combines GROVER molecular graph embeddings, ESM protein language-model embeddings, and CTD physicochemical descriptors, then exchanges drug-target information through bidirectional cross-modal attention. A shared pair representation is used for continuous affinity regression, high-affinity binary classification, and six-level affinity classification. Experiments on Davis and KIBA cover random 80/20 and DeepDTA-style standard splits. On the random 80/20 split, DPTM-DT achieves MSE/CI values of 0.193/0.917 on Davis and 0.120/0.918 on KIBA. It also reports binary AUPR/MCC values of 0.727/0.654 and 0.798/0.689, and six-class Macro-F1/Top-2 values of 0.800/0.932 and 0.815/0.962 on Davis and KIBA, respectively. Across the reported regression, binary classification, and multiclass classification settings, DPTM-DT achieves the best overall performance among the compared methods. Results under the standard split show the same relative trend. Ablations indicate that dual target representation, gated fusion, and cross-modal attention each contribute to the final performance. Code and supplementary materials are available at: anonymous.4open.science/r/DPCM-DT-74E0.

Let $a_1,\ldots,a_m$ be distinct elements of a field $K$, and let $Z(X)=\prod_{i=1}^m (X-a_i)$. We study the arithmetic complexity of computing the unique normalized Bezout pair $s,t$ satisfying $sZ+tZ'=1$, with $\deg t<m$ and $\deg s<m-1$. The classical product-tree approach requires $O(M_K(m)\log m)$ field operations, where $M_K(m)$ denotes the cost of multiplying degree-$<m$ polynomials over $K$. Thus, even when $M_K(m)=O(m\log m)$, the resulting bound is $O(m\log^2 m)$ rather than $O(m\log m)$. Over an infinite field of characteristic zero, we prove that, in the generic rational straight-line-program model, computing all coefficients of the canonical Bezout pair is equivalent, up to an additive $O(M_K(m))$ cost, to arbitrary-node multipoint polynomial evaluation and to interpolation. The main ingredient is an explicit differential reconstruction that recovers $Z$ from $(s,t)$ in $O(M_K(m))$ arithmetic operations on a nonempty Zariski-open subset. Combining this reconstruction with automatic differentiation and transposition yields the complexity equivalence. We further transfer Strassen's lower bound for the elementary symmetric functions to the Bezout problem, obtaining an $\Omega(m\log m)$ nonscalar lower bound. Hence an $O(m\log m)$ algorithm, if it exists in this model, would be asymptotically optimal. The reconstruction is genuinely characteristic-dependent: in characteristic $p$, all squarefree polynomials $X^p+cX+d$, with fixed $c\ne 0$, have the same normalized Bezout pair $(0,c^{-1})$. Therefore the all-field, all-input $O(m\log m)$ problem remains open; generically in characteristic zero, however, it is reduced to the corresponding arbitrary-node multipoint evaluation and interpolation problem.

Quantum optimization provides a promising approach for solving large-scale combinatorial problems through quadratic unconstrained binary optimization (QUBO) formulations. However, integrating QUBO-based solvers into structured optimization frameworks while preserving solution guarantees remains a fundamental challenge. This paper develops a hybrid quantum-classical column-and-constraint generation (QCCG) framework for solving two-stage adaptive robust optimization problems with binary first-stage decisions and linear recourse under polyhedral uncertainty. The proposed approach reformulates the restricted master problem as a QUBO and solves it approximately using a quantum optimizer, while retaining a classical adversarial subproblem to compute worst-case recourse and certify solution quality. We construct a constraint-preserving QUBO encoding for inequality-constrained master problems using slack variables and penalty terms, enabling general mixed-integer structures to be mapped to quantum-compatible representations. To address inexactness arising from discretization, penalty modeling, and quantum optimization, we introduce a bound-adjustment mechanism that yields valid lower and upper bounds and provides a certified stopping criterion. We show that the proposed framework generalizes classical column-and-constraint generation and retains its convergence properties when the master problem is solved exactly. Numerical experiments on two-stage robust location-transportation problems demonstrate that the proposed hybrid approach achieves solution quality comparable to classical methods while reducing the computational burden associated with solving mixed-integer master problems, highlighting the potential of hybrid quantum-classical optimization for scalable decision-making under uncertainty.

Enterprise analytics agents are not only text-to-SQL systems. They interpret business intent and choose metric definitions. They select data sources, execute tools, inspect results, and produce natural-language answers. Those answers may influence operational, financial, or executive decisions. Grading final answers hides where these agents fail. A plausible answer can use the wrong source of truth. It can skip a required decomposition, claim causality without support, or change its table interpretation across repeated runs. We present an end-to-end evaluation methodology for analytics agents. The methodology grades agent behavior across three families: semantic understanding, execution quality, and reliability. Grading uses question banks with human-written golden answers, repeated runs, and runtime traces. Each run first passes a run-validity check, then receives tiered, abstention-aware scores that feed a decision framework rather than a release gate. We instantiate the methodology on a controlled internal analytics agent at a large online marketplace. The case study uses 50 analytics questions, two anonymous model configurations, and three randomized repetitions per configuration, yielding 300 traces. The higher-capability configuration reduced early refusal from 73% to 0% and increased real-data answers from 21% to 73%. However, it also exhausted the tool-round budget on 16% of runs. It overran the schema-exploration budget on 77% of traces. It changed its table interpretation on 41 of 50 questions. On finance questions with structured golden answers, source table use and escalation improved, but canonical decomposition remained weak in both configurations. These results show why trust in analytics agents requires end-to-end, trace-backed evaluation rather than SQL correctness or final answer quality alone.

Diffusion models have become the dominant family of generative models in the visual domain. However, their widespread public availability enables misuse at scale, motivating a rapidly growing body of research on adversarial attacks and defenses. This survey provides, to our knowledge, the first unified review of this literature across three visual modalities: image, video, and 3D. We introduce a comprehensive, task-centric taxonomy: we first divide the literature by modality; within each modality, we separate methods into attacks and defenses, and then group them by the generative task they target, presenting them chronologically within each task. Moreover, we provide an in-depth analysis of their evaluation settings, consolidating the datasets, metrics, and benchmarks used to assess them. We conclude by identifying several open challenges and outlining concrete future research directions. Project Webpage: https://github.com/ozgurkara99/awesome-adv-attack-defense-on-diffusion

We study the repeated allocation of a single indivisible resource among $n$ strategic players. Each player $i$ has a privately known value distribution $D_i$, and values are drawn independently across players and periods. The goal is to find fair and efficient mechanisms. We apply the repeated first-price auction with equal initial endowments of virtual money. We show that each player can asymptotically secure the same fair-floor guarantee $f(D_i)$ as in Csóka 2026; consequently, the mechanism is $1.283$-optimal. This provides a simpler and more robust alternative mechanism for this special case and may also help derive sharper upper bounds on the price of anarchy.

This technical report studies the local weak observability of a planar bicycle-model vehicle when vehicle pose, planar LiDAR extrinsic calibration, and steering-angle bias are estimated jointly. A Lie-derivative-based nonlinear observability analysis is used to examine stationary, straight-line, constant-curvature, and combined straight-plus-arc motion. The resulting observability matrices and nullspaces describe how pose, LiDAR translation and yaw offsets, and steering bias become coupled under different motion primitives. Stationary motion and individual motion primitives retain unobservable directions, whereas the combination of straight and curved motion removes the identified degeneracies and yields full local weak observability of the seven-state system. The analysis provides a theoretical basis for selecting calibration trajectories that sufficiently excite both steering and sensor-extrinsic parameters.

An avatar that holds a conversation should decide what to say and to move while saying it, yet these abilities live in separate model families: spoken dialogue models produce speech without motion, and co-speech motion models produce motion only from audio handed to them. The standard remedy is a cascade that first generates the spoken response and then runs a motion model over the finished audio, which requires a second full inference pass and precludes any joint optimisation between the two. We present Motion-Omni, an end-to-end framework in which a spoken dialogue model natively outputs explicit facial expression together with hand, upper-body and lower-body motion, generated directly from the hidden states that produce the speech. Joint training is not optional here: with the speech pathway frozen, motion remains misaligned with the audio, and co-adapting the LLM, Speech Generator and Motion Generator under both objectives is what recovers alignment while retaining spoken-dialogue ability. Supervision comes from a scalable, model-agnostic pipeline that pseudo-labels consistent-voice speech responses with a replaceable motion teacher, yielding 422,856 quality-ranked pairs (1,402 hours). We further release SwDA-500 and, to our knowledge, the first public evaluation protocol for stochastic open-ended full-body spoken dialogue, matching audio across motion systems while unifying rendering, automatic metrics, human evaluation, and latency measurement. Instantiated with a Qwen2.5-7B-Instruct backbone, Motion-Omni-Q7 matches the same-audio teacher cascade to within 2% on reference-free motion metrics while responding 5.4 x faster (RTF=0.78, faster than real time), surpasses all non-teacher cascades on beat correlation and diversity, and reaches a 2.62% word error rate, the lowest among the omni-modal systems compared.

Existing audio-video generation methods produce well-synchronized clips but are limited to short durations, while long-video generation methods extend duration through chunk-based iterative synthesis yet lack audio entirely. Generating long audio-video jointly is fundamentally harder than either task alone: each chunk must simultaneously maintain video temporal coherence, audio temporal coherence, and cross-modal synchronization, whose conditioning signals enter the model through different pathways. In this work, we present Encore for long-form synchronized audio-video generation. Our key insight is to factor this challenge into: (1) local continuity handled by iterative generation with explicit cross-chunk context propagation, and (2) global consistency enforced via reference audio-video signals with shifted position embedding. Building on this design, we propose Adaptive Signal Routing (ASR), which introduces learnable attention biases within self-attention and learnable residual scales on cross-attention outputs, enabling the model to adaptively modulate the influence of each conditioning signal. Trained end-to-end for joint audio-video generation, Encore also supports infinite-length audio-to-video and video-to-audio synthesis at inference by conditioning on the ground-truth modality throughout the denoising process. Experiments on our extended VerseBench for long audio-video evaluation demonstrate that Encore significantly outperforms existing methods in both generation quality and temporal coherence. Code and data for this paper are at https://github.com/shaohua-pan/Encore.

Gear digital twins, hardware-in-the-loop rigs, and active vibration control close a loop around the instantaneous tooth-contact state, demanding a solver that is real-time, deterministic, and embeddable – properties that loaded tooth contact analysis (LTCA), a data-dependent linear complementarity problem (LCP) costing seconds per mesh phase in fp64, structurally lacks; learned surrogates infer fast but spend one full LTCA solve per training sample and offer no guarantee beyond their training envelope. We instead compress the closed-form contact physics – contact-point localization, principal-curvature extraction, and the elliptical Hertz solution – directly into a branch-free fixed-point FPGA datapath, termed physics-direct, realized through the fully open-source openXC7 flow (no vendor tools, no floating-point IP) and validated on retired XC7K480T silicon. Four SoCs – geometry, Hertz, and two fused paths – pass bit-exact JTAG readback against a golden model. The on-chip preview tracks LCP body pressure to within -22% and loaded transmission error (LTE) peak-to-peak to -19%, preserving the exact $W^{1/3}$ law; with zero training it extrapolates in load more accurately than a trained network (18.9% versus 20.3%). Latency is a compile-time constant ($\sigma = 0$ jitter), and the DSP-bound geometry kernel saturates the fabric near 63 lanes. Physics-direct is an interpretable, deterministic alternative to neural surrogacy for embedded tooth-contact estimation.