SEARCH · Searcharxiv

Search Searcharxiv

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 307 records · Page 17Linked to original sources

arXiv2026

Exploiting Overlapping Fields of View for Redundancy-Aware Uplink Transmission in Vehicular 6G

Emerging uplink-dominant 6G use cases, such as cooperative vehicular streaming, require efficient transmission of high-volume visual data over limited wireless resources. While semantic communications can reduce traffic by prioritizing task-relevant content, most existing approaches treat users independently and therefore overlook spatial redundancy among nearby devices' observations. This paper proposes a semantic-aware multiple access scheme that exploits overlapping fields of view among vehicular users to reduce redundant uplink transmissions. We formulate a joint perception and transmission control problem in which users decide which image patches to transmit, when to transmit them, and over which channel, subject to communication constraints. To address the resulting complexity, we introduce a practical two-phase approach. First, nearby vehicles share selected observation patches over Vehicle-to-Vehicle (V2V) links to calculate inter-user spatial redundancy. Second, users transmit only semantically important, non-redundant patches to the base station, where observations can be reconstructed using the received patches and complementary views from neighboring vehicles. Simulation results in a dense urban vehicular scenario demonstrate that our approach improves the proportion of users who achieve high-fidelity reconstruction, highlighting the potential of semantic-aware multiple access for sustainable and resource-efficient 6G uplink systems.

cs.NI↗

arXiv2026

Decomposing Wage Stagnation: Employment Reallocation, Wage Structure,and Demographics

Japan's average log real hourly wages rose until the mid-1990s, declined through the mid-2010s, and partially recovered thereafter. This paper decomposes these changes over 1980-2024 into four components: demographic change across worker types, changes in relative employment shares across job types, changes in relative log wages across job types, and unweighted mean wage growth. The framework combines a shift-share decomposition across worker types with an extension of the Olley-Pakes decomposition across job types within worker types, separating employment reallocation from changes in relative wage structure. The four components vary across periods. Before 1996, unweighted mean wage growth and changes in relative wage structure contribute positively, while demographic change and employment reallocation contribute negatively. During 1996-2014, all four components are negative. After 2014, the recovery mainly reflects unweighted mean wage growth. Employment reallocation and changes in relative wage structure contribute differently across dimensions of job type.

econ.GN↗

arXiv2026

Conditional Co-Ablation: Recovering Self-Repair Backups in Transformer Circuits

Mechanistic interpretability seeks to explain transformer behavior through circuits: sets of internal components that causally support a behavior. However, self-repair creates a blind spot: ablating a primary component can activate a dormant backup, so a circuit that explains behavior in the intact model can become incomplete under the intervention used to test it. We formulate this gap as conditional circuit completion: given a primary set, identify components that become causally important after its removal. We introduce conditional co-ablation (CoAx), which ranks candidates by growth in ablation effect after primary-set removal. We show that a perfectly dormant backup can be indistinguishable from an irrelevant component to per-unit intact-state scores, whereas its conditional effect change exactly aggregates all interaction orders linking it to the removed set. On GPT-2-small's Indirect Object Identification (IOI) circuit, CoAx recovers the documented backup heads at 0.941 ROC-AUC, versus 0.815 for the strongest intact-state attribution baseline and 0.758 for the matched conditional-energy control. Recovery drops to 0.40 +/- 0.13 AUC for alternative component sets matched in behavioral effect, output displacement, and depth, showing that recovery is specific to the removed circuit. Beyond recovery, the CoAx-selected heads are causally load-bearing: freezing them after primary removal sharply reduces the IOI margin, while adding them to the incomplete circuit reduces incompleteness from 0.75 to 0.21. More broadly, conditional growth aligns with intervention-derived repair in 11/12 held-out instances across 4 mechanism clusters, and CoAx completions outperform matched random completions on all 8 non-GPT-2 models spanning 6 architecture families. Together, causal explanations of self-repairing transformers must account for backup circuitry when primary components fail.

cs.LG↗

arXiv2026

A rubric-based controlled comparison of frontier language models on expert-authored clinical reasoning tasks

Multiple-choice medical benchmarks are increasingly saturated, and recent rubric-based evaluations such as HealthBench have shown that open-ended clinical performance is far from solved - its "Hard" subset top score remains 32%. We present a small, deliberately difficult evaluation dataset of five clinician-authored clinical scenarios spanning four specialties (anaesthesia, internal/family medicine, emergency medicine, and obstetrics), each accompanied by an atomic, weighted, MECE rubric (25-62 criteria per task; 184 criteria total) authored from a clinician-drafted golden answer. We evaluate three frontier models: GPT 5.4, Claude Opus 4.7, and Gemini 3.1 Pro. Mean rubric pass rates were 0.47 (Claude), 0.38 (GPT), and 0.37 (Gemini). The central finding is an inversion of clinical priority: the highest-weighted (weight-5, critical) criteria passed at only 32.4-41.7%, while low-stakes weight-1 criteria passed at 80-90%. 55 of 108 critical (weight-5) criteria (51%) were satisfied by no model. Three LLM autoraters reproduced expert met/not-met labels on 92.8-94.6% of 552 graded criteria. We position this as a methods-and-preliminary-findings contribution: the five tasks demonstrate a scalable, defensible pipeline ready to develop into a large-scale benchmark.

cs.AI↗

arXiv2026

Bridgeland-Enriques general K3 surfaces

This article introduces a notion of Bridgeland-Enriques general K3 surfaces motivated by the study of Enriques categories over K3 surfaces and the invariant Bridgeland stability conditions. The family of Bridgeland-Enriques general K3 surfaces of degree 10 detects a categorical degeneration of special Gushel-Mukai threefolds. Also, the families of Bridgeland-Enriques general K3 surfaces with higher degrees are closely related to Hodge-special Gushel-Mukai fourfolds and double EPW sextics.

math.AG↗

arXiv2026

A Constructive Framework for Generalized Fourier Transforms via Truncate-and-Generalized Limits

This paper introduces a constructive definition of generalized Fourier transforms based entirely on ordinary truncated Fourier integrals and ordered dual-domain limits, within the framework of improper Riemann integration and classical analysis. The proposed truncate-and-generalized-limit (t.g.l.) formulation does not require test-function spaces, Lebesgue measure theory, or duality pairings in its proofs: the forward and inverse transforms are defined directly through finite-domain truncation of the target function, followed by successive ordered limits in the time and frequency domains. As consequences of this constructive definition, the formulation provides a unified treatment of non-decaying, oscillatory, and locally singular functions beyond the classical L1(R) setting; reveals an inherent asymmetry between the forward transform, interpreted as a first-order generalized-limit family, and the inverse transform, which requires frequency-domain truncation to generate pointwise reconstruction through Dirichlet-type oscillatory localisation; and clarifies the distinction between the t.g.l. approach and distribution theory, where generalized Fourier transforms are introduced through duality pairings rather than constructed from ordinary integrals. The inversion formula is established rigorously for two concrete admissible classes using only the classical Dirichlet convergence theorem. Several examples confirm that the framework covers constants, polynomials, periodic functions, singular kernels, and chirp signals within a single constructive scheme.

math.FA↗

arXiv2026

Anatomy of the Quasi-PDF with a Transverse-Momentum Cutoff

The quasi-PDF is fundamentally different from the light-cone PDF, and large-momentum effective theory relates the two through perturbative matching. This has required considerable effort. One of the difficulties in matching is that no single renormalization scheme has been agreed upon for the quasi-PDF, and the resulting variety of prescriptions can obscure the physics underlying the matching. We study how the choice of scheme affects the renormalized quasi-PDF and the matching coefficient, and how the two are related, aiming to identify a scheme in which this dependence becomes transparent. As a first attempt, we consider the transverse-momentum cutoff scheme. It is not used in practice, since a hard momentum cutoff breaks Lorentz invariance, and it has generally been treated as an illustrative example exhibiting the linear divergence of the spatial Wilson line. A careful study, however, reveals a richer structure than expected. The ultraviolet divergence in this scheme depends not only on the cutoff itself, but also on an independent divergence associated with the momentum fraction of the quasi-PDF. Disentangling these two ultraviolet structures, and following the disentangling through the explicit one-loop calculation, uncovers further structure along the way. Once the ultraviolet contributions are correctly separated and assigned to the renormalization counterterm, a finite ambiguity remains, and we use it to restore the quark-number conservation broken by the cutoff. The resulting renormalized quasi-PDF and matching coefficient can be meaningfully compared with the result obtained in dimensional regularization, and we show how the two are reconciled.

hep-ph↗

arXiv2026

An Equivalence result for sketched Anderson Acceleration and sketched GMRES

In this paper we present an equivalence result between a randomized version of Anderson Acceleration and of randomized GMRES for linear problems. Namely, we extend the classical result of Walker and Ni (2011) to the case in which the least-squares problem in Anderson Acceleration is solved in a sketched space defined by a random projection. This equivalence suggests potential avenues for further research in the design and analysis of randomized acceleration methods.

math.NA↗

arXiv2026

When do prophets profit in prediction markets?

Prediction markets aggregate dispersed beliefs into prices that act as probabilistic forecasts of uncertain events. Classical theory establishes how a better-than-market forecast can yield positive trading profit. However, it hinges crucially on the specific automated market maker (AMM) design, and is not applicable to popular exchanges today which are based on central limit order books. This paper fills that gap. For any prediction market and any proper scoring rule $S$, we exhibit a ``proper'' betting strategy that depends only on the forecaster's prediction $\mathbf{p}$ and the market price $\mathbf{q}$, and earns positive expected profit \emph{whenever} $\mathbf{p}$ outperforms $\mathbf{q}$ under $S$ and the market has sufficient liquidity. Moreover, this proper betting is essentially the only strategy with such robust profitability guarantee. Our proof rests on a decomposition of expected profit that strictly generalizes the classical AMM guarantee and also explains how strategies can profit even without an accuracy edge. Empirically, across thousands of forecasts by AI models, proper betting is the only strategy that reliably converts accuracy into profit, and we further identify systematic forecasting personas and show how the optimal proper strategy varies across them. For feasibility demonstration, we run a monthlong live pilot test on Kalshi; the encouraging preliminary results show that proper betting can survive real-world spreads, fees, discrete fills, and limited liquidity.

cs.AI↗

arXiv2026

SCI-Mamba: Unsupervised Learning based Low-Light Image Enhancement for Non-Cooperative Spacecraft

Low-light visual perception acts as the core visual foundation for on-orbit servicing missions targeting non-cooperative spacecraft, supporting autonomous rendezvous, pose estimation, component detection and robotic capture operations. Spaceborne imagery suffers from severe low-light degradation, while the extreme scarcity of paired normal/low-light space samples severely limits the generalization capacity of supervised enhancement algorithms. To address this practical bottleneck, this paper proposes SCI-Mamba, an unsupervised enhancement network for low-light orbital spacecraft observations. The proposed framework unites self-calibrated unsupervised learning, linear-complexity VMamba architecture and Retinex physical priors, delivering a lightweight enhancement pipeline adaptable to resource-limited spaceborne hardware. We construct Space Dark-1.0, a dedicated low-light spacecraft dataset integrating real orbital footage, darkroom hardware-in-the-loop measurements and physically constrained synthetic data covering diverse illumination, motion and attitude conditions. Comprehensive comparisons with CNN-, Transformer- and prevailing Mamba-based approaches verify the advantages of SCI-Mamba in visual authenticity, color fidelity and inference speed. The proposed framework provides a practical low-light enhancement solution for close-proximity non-cooperative space operations. The code is available at https://github.com/bitswh/SCI-Mamba

eess.IV↗

arXiv2026

IB-Flow: Information Bottleneck-Guided CFG Distillation for Few-Step Text-to-Image Generation

While large-scale text-to-image generative models have achieved unprecedented visual performance, their inherent reliance on multi-step iterative solvers incurs severe inference latency. Few-step distillation targeting the Classifier-Free Guidance (CFG) trajectory has emerged as the prevalent dual-dimensional compression paradigm. However, existing frameworks remain subjugated by a coarse-grained blind injection paradigm that perpetually enforces a globally static guidance strength while indiscriminately sampling the supervisor timestep. This state-agnostic design completely disregards the intrinsic nature of image generation as a dynamic evolutionary process characterized by progressive entropy reduction, which not only restricts the performance boundary of few-step compression but also precipitates severe CFG over-conditioning artifacts. To transcend these limitations, we re-examine the distillation procedure through the theoretical lens of Information Theory, formally modeling it as a dynamic mutual information game constrained by the Information Bottleneck (IB) principle. Specifically, we dismantle traditional blind assumptions via a dual-track adaptive framework. To determine the injection target, we propose an instance-aware selection mechanism that transmutes the intractable KL divergence constraint into a zero-overhead closed-form solution predicated on the local vector field norm. To regulate the injection strength, we introduce an entropy-aware schedule that dynamically decays alongside the SNR, applying maximal thrust for initial structural anchoring before smoothly reverting to the natural manifold to refine micro-details. Extensive empirical evaluations corroborate that our framework fundamentally eradicates over-conditioning artifacts, shattering the performance ceiling to achieve SOTA generative fidelity under extremely stringent 2-step configurations.

cs.CV↗

arXiv2026

Quantum Approximate Optimization via Noise-Directed Adaptive Warm-Starting

Progress towards a quantum advantage using known heuristic methods for combinatorial optimization is impeded by hardware noise and limited qubit count. Here, we propose a noise-aware adaptive approach to quantum approximate optimization, Noise-Directed Adaptive Warm-Starting (ND-AWS), that builds on recent concepts such as Warm-Start QAOA and Noise-Directed Adaptive Remapping. By leveraging bitflip gauge transformations, our algorithm exploits amplitude-damping-like noise components. We experimentally implement high-performance quantum optimization ansätze on 100-qubit Ising Hamiltonians, showing that ND-AWS generally improves the performance over a non-gauge-transformed iterative Warm-Starting variant, at no additional circuit cost. This places our results among the highest-quality demonstrations of quantum optimization with similar ansätze at this scale. Crucially, the simplicity of the framework opens the door for future enhancements such as adaptive bias schedules, and integration with classical solvers.

quant-ph↗

arXiv2026

Dynamics of Biased Domain Walls: The Rocket Effect

We investigate the dynamics of domain walls in scalar field theories with degenerate vacua (i.e., vacua of equal energy density) in which the scalar field mass depends on the vacuum state. Using analytical arguments and numerical simulations, we show that this vacuum dependence of the scalar field mass renders the emission of scalar radiation from domain walls anisotropic, preferentially toward regions with smaller scalar field mass. We further show that the resulting recoil (rocket) effect biases the evolution of cosmological domain wall networks in favor of the lower-mass vacuum, thereby promoting network decay. We also demonstrate that the biased evolution of domain walls in theories with degenerate vacua, previously attributed to asymmetries of the potential barrier near the local maximum, is instead primarily controlled by the vacuum dependence of the scalar field mass. More generally, in theories with non-degenerate vacua, this recoil mechanism constitutes an additional source of dynamical bias that can either hasten or delay network decay relative to the standard expectation based solely on differences in vacuum energy density.

astro-ph.CO↗

arXiv2026

Energy-guided Recursive Model

Recursive models show promise on reasoning and language tasks, yet their test-time scaling lacks a principled criterion for selecting trajectories or determining recurrent depth. We introduce \textbf{Energy-guided Recursive Model (ERM)}, which uses Hopfield-type memories of valid local and global structures to assign intrinsic energies to candidate trajectories. These energies guide candidate selection and suggest an effective range of recurrent depths, implying that deeper recurrence does not necessarily improve reasoning accuracy. They also enable sampling methods such as parallel tempering to improve exploration. For reasoning tasks, ERM achieves optimal solutions on Sudoku ($98.97\%$), Pencil Puzzle Bench (PPBench, $88.04\%$) and Maze ($99.30\%$), reaching the best accuracy in recursive modeling. On language modeling, ERM reduces RedPajama-V2 perplexity by $1.74\%$ with marginal inference overhead. The results support energy guidance as a practical framework for improving test-time scaling in recursive models.

cs.LG↗

arXiv2026

tidyHEBO: Robust General-Purpose Bayesian Optimization with Model-Consistent Warping and Pareto Search

Bayesian optimization (BO) is widely used for expensive black-box problems, yet practical performance depends not only on high-level algorithmic choices but also on how surrogate model training, input and output warping transformations, acquisition functions, and candidate search are implemented. We present tidyHEBO, a BoTorch-native single-objective optimizer designed for robust general-purpose optimization. tidyHEBO jointly fits Yeo-Johnson output warping with the Gaussian-process surrogate, evaluates acquisition functions on the original objective scale using deterministic quadrature or MC-samples, and performs constrained cumulative Pareto search over multiple acquisition criteria. Without any Olympus-specific hyperparameter tuning - using only default optimizer configurations - tidyHEBO ranked first among the evaluated methods on the Olympus benchmark. It achieved the best average ranks for typical performance (average rank 1.53), worst-tail performance (1.21), and run-to-run variability (2.00), measured by median nAUC, CVaR_nAUC, and IQR_nAUC, respectively. Using the same default configuration, tidyHEBO also performed strongly on synthetic and Needle-in-a-Haystack problems and closely matched HEBO on Bayesmark (92.64 versus 93.34) while exceeding GP with logarithmic expected improvement and random search. Adaptive batching reduced feedback rounds while revealing a controllable trade-off between parallelization and optimization quality as the batch cap increased. These results characterize tidyHEBO as a robust, reproducible general-purpose optimizer for a broad range of practical optimization problems, including scientific applications and hyperparameter tuning.

cs.LG↗

arXiv2026

Same Stories, Different Journeys: Exploring Persona-Grounded Conversational Agents for Supporting Career Exploration with Peers' Posts

Young job seekers frequently explore their career possibilities by browsing peers' posts that share job-seeking experiences. However, static browsing requires them to reconstruct fragmented cases and privately judge what others' experiences mean for themselves, sometimes intensifying anxiety through upward social comparison. In this paper, we examine how transforming these posts into persona-grounded conversations reshapes this sensemaking process. We developed JobMate, a prototype featuring agents that have personas built upon peers' posts and follow the self-determination theory to converse with users. In a between-subjects comparative study with 24 participants, RedNote browsing exposed diverse trajectories but left reconstruction and comparison largely to users, whereas JobMate supported case selection and continued questioning. The conversations further prompted users to articulate previously implicit constraints and accept, challenge, or revise the agent's interpretations. We discuss design implications for combining authentic peer experiences with generative AI in career exploration.

cs.HC↗

arXiv2026

Omni-Decision: Evidence-Ledger Planning for Omni-Modal Agents

Omni-modal agents must seek evidence across video, audio, web pages, and computation to answer questions. Their main bottleneck is planning: noisy multimodal observations accumulate in conversation history and disrupt later decisions, while multimodal models have limited capacity for multi-step planning. Controlled backend replacements support this diagnosis: replacing the planner causes a much larger performance loss than replacing the perception backend. We present Omni-Decision, an omni-modal agent built on evidence-ledger planning: it replaces the growing dialogue history with an explicit evidence ledger that records what evidence is still missing, what has been confirmed, and where records conflict. A critic reads each noisy observation and passes only the usable content to the ledger, discarding the rest, so the planner works from a compact context throughout the task. Each run records the state, action, and verdict at every step, and supervised fine-tuning and decision-level reinforcement learning on these trajectories further improve the planner. Omni-Decision achieves state-of-the-art accuracy of 81.4% on OmniGAIA at approximately 43% of Gemini-3.1-Pro's cost per question, and 65.0% on WorldSense long-video understanding, level with the strongest end-to-end model.

cs.AI↗

arXiv2026

Turán number of a matching and a Berge triangle

For a fixed graph $G$, an $r$-uniform hypergraph is said to contain a Berge-$G$ if there exists a bijection $f\colon E(G)\to E(\mathcal{H})$ for some subhypergraph $\mathcal{H}$ such that $e\subseteq f(e)$ for every $e\in E(G)$. Motivated by Alon and Frankl's study of Turán problems under bounded matching constraints, we investigate the maximum number of edges in $r$-uniform Berge-$K_3$-free hypergraphs with matching number at most~$s$. We determine the exact Turán numbers for the cases $r=3$ and $r=4$. For $r=3$ and $n \geq 3 s$, we prove that every $n$-vertex Berge- $K_3$-free 3-graph with matching number $s$ has at most $s(n-2 s)$ edges, and we characterize the unique extremal hypergraph attaining equality. For $r=4$ and $n \geq 4 s$, the maximum number of edges is $s\lfloor(n-2 s) / 2\rfloor$, except for the exceptional case $s=1$ and $n \equiv 1(\bmod 4)$, in which the bound is $(n-1) / 2$. As a corollary, our results recover the classical theorem of Győri on Berge-$K_3$-free hypergraphs.

math.CO↗