Expander SAEs apply left-d-regular expander masks to TopK SAEs, learning only dn decoder parameters instead of mn and tracing a storage-fidelity frontier that reaches 293x compression with 84% retained performance on Qwen2.5-3B.
Canonical reference
A graph-based recovery and decomposition of Swanson’ s hypothesis using semantic predications
Canonical reference. 80% of citing Pith papers cite this work as background.
1095
Pith papers citing it
1
external citations · Crossref
Background
80% of classified citations
citation-role summary
background 254
method 24
baseline 6
dataset 5
extension 2
citation-polarity summary
authors
co-cited works
representative citing papers
The normalized orbit of a bounded normal operator can be a frame, providing a counterexample to Conjecture 3.
Introduces the largest freely available Italian clinical notes corpus with 4M notes and expert-annotated subset for a new CRF-filling benchmark.
Machine learning methods discover a new noncrossing-partition statistic interpreting q,t-Narayana polynomials and yield a combinatorial proof of their symmetry.
SLayerGen generates crystals invariant to any space or layer group via autoregressive lattice and Wyckoff sampling plus equivariant diffusion, achieving gains over bulk models on diperiodic materials after correcting a prior loss inconsistency for hexagonal groups.
Every proper minor-closed graph class admits an optimal (1+o(1)) log n bit adjacency labeling scheme.
A directed weighted two-graph model separates feasibility from movement in solution discovery and yields a detailed complexity classification for path and shortest-path discovery.
The method reformulates ALE mesh motion as independent multi-patch spline parameterizations per time step, using barrier functions, tangential-slip reparameterization, and constant-preserving quasi-interpolation to enable large-rotation FSI simulations.
Superconductivity in high-pressure MnB4 is induced by altermagnetic spin fluctuations, yielding extended-s pairing symmetry.
A new qubit-efficient HUBO encoding for graph partitioning problems like minimum coloring uses logarithmic bits and a lexicographic penalty to cut resources while providing provable optimality conditions.
A survey of 172 open educational datasets from 204 papers across LAK, EDM, and AIED conferences reveals trends, 143 previously uncatalogued datasets, field gaps, and an 8-item PRACTICE checklist for better data publication.
A microlocal lift of Navier-Stokes dynamics on manifolds yields an if-and-only-if geometric criterion for solution blow-up in terms of deformation integrability, directional entropy, and lifted energy.
A 9U CubeSat detector can identify a thermonuclear weapon on a satellite from 4 km away by observing spallation neutrons induced by GeV protons in roughly one week.
O(n log n) algorithm and matching Omega(n log n) lower bound for partitioning a simple polygon's boundary into the minimum number of contiguous visible segments.
Introduces a method to design structure-specific relational inductive biases for a base transformer architecture, enabling end-to-end transcription of documents with intrinsic structures, demonstrated on sheet music, shape drawings, and mechanical engineering drawings.
The paper introduces a probabilistic sign rule for quotients of positive series and integral transforms that reduces monotonicity, log-supermodularity, and log-convexity to kernel sign criteria via moment identities, and applies it to derive new inequalities for hypergeometric, Stieltjes, and Prabha
SPoILeR uses multimodal pre-training to enable accurate novel view synthesis of infrared, polarimetric, and multispectral data from RGB-supervised fine-tuning on new scenes.
NEvo performs evolutionary search guided by a dynamic voxel-level encoding model to synthesize videos that maximize predicted activity in target brain ROIs, recovering known selectivities and revealing temporal dynamics differences.
Every n-vertex H-minor-free graph admits a 3-coloring with monochromatic components of size O_H(n^{4/9}).
The Spin-MInt algorithm is proven symplectic for general K electronic states via explicit verification of the condition MJM^T = J on the coadjoint orbit of the su(K) Lie-Poisson algebra.
The authors synthesize a typology of fourteen OSS sub-genres from a review of 3,925 papers and present a research agenda on cross-sub-genre generalization.
Structural identifiability analysis shows point sources restore identifiability for inferring spatial stochastic dynamics parameters from static snapshots, unlike distributed sources, with limits depending on modeling choices.
Maximal quantum leakage upper-bounds quantum inference accuracy; optimal encodings are pure states, with tight frames and equiangular tight frames optimal when system dimension is small.
In multistage SI(k)R models, the relationship between prevalence peak and weighted stage functional maxima varies with scaling of progression rates, converging under Erlang scaling to a delay model that justifies the factor-two approximation with error bounds and corrections.
citing papers explorer
-
SLayerGen: a Crystal Generative Model for all Space and Layer Groups
SLayerGen generates crystals invariant to any space or layer group via autoregressive lattice and Wyckoff sampling plus equivariant diffusion, achieving gains over bulk models on diperiodic materials after correcting a prior loss inconsistency for hexagonal groups.
-
Adjacency labelling for proper minor-closed graph classes
Every proper minor-closed graph class admits an optimal (1+o(1)) log n bit adjacency labeling scheme.
-
Parameterization-driven arbitrary Lagrangian-Eulerian method for large-deformation isogeometric fluid-structure interaction
The method reformulates ALE mesh motion as independent multi-patch spline parameterizations per time step, using barrier functions, tangential-slip reparameterization, and constant-preserving quasi-interpolation to enable large-rotation FSI simulations.
-
Superconductivity induced by altermagnetic spin fluctuations in high-pressure MnB$_4$
Superconductivity in high-pressure MnB4 is induced by altermagnetic spin fluctuations, yielding extended-s pairing symmetry.
-
Qubit-efficient and gate-efficient encodings of graph partitioning problems for quantum optimization
A new qubit-efficient HUBO encoding for graph partitioning problems like minimum coloring uses logarithmic bits and a lexicographic penalty to cut resources while providing provable optimality conditions.
-
Open Datasets in Learning Analytics: Trends, Challenges, and Best PRACTICE
A survey of 172 open educational datasets from 204 papers across LAK, EDM, and AIED conferences reveals trends, 143 previously uncatalogued datasets, field gaps, and an 8-item PRACTICE checklist for better data publication.
-
NeuroTrain: Surveying Local Learning Rules for Spiking Neural Networks with an Open Benchmarking Framework
A taxonomy of SNN training algorithms is presented with the release of NeuroTrain, an open benchmarking framework for reproducible comparisons across datasets and architectures.
-
What Should Explanations Contain? A Human-Centered Explanation Content Model for Local, Post-Hoc Explanations
A 14-code content model for local post-hoc AI explanations, derived from 325 user statements and validated by experts with high reliability scores.
-
Asymptotically Optimal Ergodic Coverage on Generalized Motion Fields
A flow-adaptive ergodic coverage formulation using MMD that preserves guarantees over evolving domains and supports open-loop planning for robots in flows.
-
Human face perception reflects inverse-generative and naturalistic discriminative objectives
Human face perception aligns with neural networks trained on inverse-generative and naturalistic discriminative tasks, as these best predict human dissimilarity judgments on controversial and random face pairs.
-
Scaling Solutions of Matter Form Factors in Asymptotically Safe Quantum Gravity
A non-trivial UV fixed point for the scalar matter form factor exists in asymptotically safe quantum gravity, with a discrete spectrum of critical exponents and infrared locality restored.
-
MulTaBench: Benchmarking Multimodal Tabular Learning with Text and Image
MulTaBench is a new collection of 40 image-tabular and text-tabular datasets designed to test target-aware representation tuning in multimodal tabular models.
-
Statistical Model Checking of the Keynes+Schumpeter Model: A Transient Sensitivity Analysis of a Macroeconomic ABM
Statistical model checking on the K+S model shows macro-financial and structural parameters produce stronger transient effects on unemployment and GDP growth than heuristic-rule parameters under fixed precision policies.
-
Inverse Design of Metainterfaces for Static Friction Control: Beyond the Hertzian Limit
A differentiable contact mechanics engine embedded in a neural network and quadratic optimizer discovers axisymmetric asperity topographies that produce target nonlinear friction laws, validated against BEM simulations.
-
Fast-Food Intimacy: How Chinese Women Navigate Soul's AI Boyfriend
Users experience fast-food intimacy with Soul's AI boyfriend that conflicts with gradual cultural expectations, introduces technical uncertainty, and shifts emotional labor onto women.
-
ProtoSSL: Interpretable Prototype Learning from Unlabeled Time-Series Data
ProtoSSL discovers generalizable prototypes from unlabeled time-series via self-supervision and assigns them to new tasks for interpretable predictions, outperforming supervised baselines in low-data regimes on ECG datasets.
-
Charge-Transfer Induced Reactivity in sp Carbon Atomic Wires: Towards 0-D sp-sp2 Nanostructures
Electrochemical reduction of hydrogen-capped polyynes yields stable amorphous sp-sp2 carbon nanoparticles with tunable diameters, >60% retained sp fraction, and ambient stability exceeding six months.
-
Raising the Ceiling: Better Empirical Fixation Densities for Saliency Benchmarking
A mixture model with adaptive KDE and per-image cross-validation raises estimated human fixation consistency by 5-15% median log-likelihood and up to 2 AUC points over fixed-bandwidth Gaussian baselines.
-
A Variational Kolosov--Muskhelishvili Network for Elasticity and Fracture
A variational physics-informed neural network using Kolosov-Muskhelishvili potentials is introduced for 2D linear elasticity and fracture problems, embedding crack conditions directly into the ansatz.
-
Dynamic Mechanical Response of Spinodal Architectures Across Length and Time Scales
Macroscale spinodal structures exhibit inertia-driven tenfold strength increase at high strain rates unlike microscale counterparts governed by constituent material sensitivity, with regime maps showing length-scale dependence analogous to fluids.
-
BadmintonGRF: A Multimodal Dataset and Benchmark for Markerless Ground Reaction Force Estimation in Badminton
BadmintonGRF is a new public multimodal dataset and benchmark that pairs multi-view video with instrumented GRF for markerless load estimation in badminton.
-
Hindered transport of spherical particles in cylindrical pores: The role of structural heterogeneity in rejection-permeability trade-offs
Coupled heterogeneity in pore and particle size distributions broadens effective Peclet number variability and systematically improves permeability at fixed rejection in membrane transport.
-
EO-Gym: A Multimodal, Interactive Environment for Earth Observation Agents
EO-Gym supplies an executable multimodal environment and 9k-trajectory benchmark that turns Earth Observation into a tool-using, multi-step reasoning task, revealing that current VLMs struggle on temporal and cross-sensor workflows while fine-tuning lifts Pass@3 from 0.49 to 0.74.
-
EnCoDe: Energy Estimation of Source Code At Design-Time
EnCoDe enables design-time prediction of block-level energy consumption in Python code via static features and ML models trained on a dataset from 18,000 programs, achieving R²=0.75 and 80.6% hotspot classification accuracy.
-
Linguistically Informed Multimodal Fusion for Vietnamese Scene-Text Image Captioning: Dataset, Graph Framework, and Phonological Attention
Introduces ViTextCaps dataset and PhonoSTFG phonological graph fusion framework for Vietnamese scene-text image captioning, showing cross-modal graph edges harm performance.
-
Compartment Modelling of Multiphase Reactors using Unsupervised Clustering
CLARA generates accurate compartment models from CFD simulations of multiphase reactors using clustering to enable faster control and design.
-
Discovery of Sparse Invariant Subgrid-Scale Closures via Dissipation-Controlled Training for Large Eddy Simulation on Anisotropic Grids
Sparse regression yields explicit invariant polynomial SGS closures for LES on anisotropic grids that achieve neural-network accuracy with simpler forms and lower computational cost.
-
EOS-Bench: A Comprehensive Benchmark for Earth Observation Satellite Scheduling
EOS-Bench creates thousands of satellite scheduling test cases spanning small to large scales and evaluates multiple solver types across five performance metrics.
-
Green-Red Watermarking for Recommender Systems
GREW uses a secret-key-driven green-red item partition and three ranking-integrated modules to embed verifiable watermarks in recommender systems that resist extraction attacks without data injection.
-
Generalized relative locality and causal sets
A fibre-bundle model with dual spacetimes (smooth manifold plus causal set) derives relative locality in general PQG theories without requiring momentum-space curvature.
-
Some results on small ordered and cyclic Ramsey numbers
Authors compute new small two-color ordered and cyclic Ramsey numbers for monotone paths, cycles, stars, complete graphs and nested matchings via SAT solving, determine closed forms for several pairs of graph classes, obtain bounds, apply reinforcement learning for lower bounds, and introduce permut
-
SCARFACE: a harmonized spatio-temporal dataset integrating socio-economic, environmental, and agricultural indicators for the Po Valley (Italy), 2011--2024
SCARFACE provides a harmonized annual panel dataset of over 2,700 socio-economic, environmental, and agricultural indicators for 256 sub-regions in the Po Valley, Italy, spanning 2011 to 2024.
-
X-VC: Zero-shot Streaming Voice Conversion in Codec Space
X-VC achieves zero-shot streaming voice conversion via one-step codec-space conversion with dual-conditioning acoustic converter and role-assignment training on generated paired data.
-
Optical Appearance of Scalarized Kerr-Newman Black Holes with Multiple Light Rings
Scalarized Kerr-Newman black holes develop an inner photon shell that produces an additional critical curve and distinct crescent-like higher-order images unlike those in Kerr spacetime.
-
Kinetic models of opinion-driven epidemic dynamics modulated by graphons
Kinetic models of opinion-driven epidemics on graphons prove L1 and Sobolev convergence, introduce a structure-preserving scheme, and show that a time-dependent reproduction number analog can generate waves without external forcing.
-
A three-dimensional morphoelastic model for self-oscillations in polyelectrolyte hydrogel filaments
A 3D morphoelastic rod model with local Stokes hydrodynamics predicts flutter instability and self-sustained 2D or 3D oscillations in clamped elliptic hydrogel filaments under constant axial electric field, with a secondary bifurcation to large-amplitude motions.
-
Existence and uniqueness of nonlocal nonlinear conservation laws via fixed-point methods
Existence and uniqueness of weak entropy solutions for nonlocal nonlinear scalar conservation laws is proven on short time horizons via fixed-point methods, extending to any finite horizon under additional assumptions.
-
Vision-Language Foundation Models for Comprehensive Automated Pavement Condition Assessment
Instruction-tuned vision-language model PaveGPT, trained on a large unified pavement dataset, achieves substantial gains over general models in comprehensive, standard-compliant pavement condition assessment.
-
Microscopic contributions to the deviation from Amontons friction law
MD simulations with ML force fields reveal non-monotonic friction-load curves in MX2/metal heterostructures arising from coexistence of longitudinal, lateral-slip, and zig-zag sliding modes.
-
Fast and accurate noise removal by curve fitting using orthogonal polynomials
Reformulating local polynomial fitting with orthogonal Chebyshev polynomials yields two algorithms that cut memory use, improve scalability, and deliver orders-of-magnitude better numerical accuracy than Vandermonde-based methods for Savitzky-Golay filters.
-
Qurator: Scheduling Hybrid Quantum-Classical Workflows Across Heterogeneous Cloud Providers
Qurator jointly optimizes queue time and fidelity for hybrid quantum-classical workflows across providers using quantum-aware DAG scheduling and a unified logarithmic fidelity score, achieving 30-75% wait reduction at high load with bounded accuracy cost.
-
A Pontryagin class obstruction for purely electric and purely magnetic Weyl curvature tensors
Vanishing of top-degree Pontryagin class products obstructs purely electric or magnetic Weyl curvature on compact 4k-manifolds.
-
Entropy correction artificial viscosity for high order DG methods using multiple artificial viscosities
Multiple artificial viscosities with analytical optimal parameters enable more flexible entropy-stable DG simulations than monolithic viscosity for 1D and 2D problems.
-
Measure charge transport in high-energy nuclear collisions with an energy scan of isobaric collisions
An energy scan of isobaric collisions provides a double-ratio method to measure electric charge transport rapidity dependence, with simulations showing exponential decrease and model-dependent slopes distinct from baryon transport.
-
Almost all graphs have no cospectral mates with height relative small to its order
Almost all graphs of order n have no cospectral mates with height o((n / ln n)^{1/10}).
-
A forward-angle large-acceptance magnetic spectrometer
A forward-angle magnetic spectrometer with 70 msr acceptance was built using a beam-passing slit in the yoke, magnetic shielding, and a counterweight for stability to support high-luminosity nuclear physics experiments.
-
Lifetime-Aware Design for Item-Level Intelligence at the Extreme Edge
FlexiFlow optimizes carbon footprint for item-level intelligence on flexible electronics by modeling lifetime variation, delivering 1.62X microarchitectural and 14.5X algorithmic reductions plus a 30.9 kHz tape-out.
-
TabPFN-3: Technical Report
TabPFN-3 scales tabular foundation models to 1M rows with synthetic pretraining, test-time compute, and benchmark-leading performance on tabular, relational, and tabular-text tasks while being up to 20x faster than TabPFN-2.5.
-
IfcLLM: Natural Language Querying of IFC Models through Complementary Relational and Graph Representations
IfcLLM combines relational and graph representations of IFC models with an LLM agent to achieve 93.3-100% first-attempt accuracy on natural language queries across three models and 30 scenarios.
-
Code-Centric Detection of Vulnerability-Fixing Commits: A Unified Benchmark and Empirical Study
Code language models show no transferable security understanding from code diffs alone, rely on commit messages, miss over 93% of fixes at 0.5% false positive rate, and suffer large drops under group or temporal splits.