Hyper Separation Logic extends separation logic and Hyper Hoare Logic with a hyper separating conjunction to support arbitrary quantifier alternation for hyperproperties over heap programs, with a soundness proof in Isabelle/HOL.
super hub Canonical reference
Scalable Funding of Bitcoin Micropayment Channel Networks
Canonical reference. 82% of citing Pith papers cite this work as background.
192
Pith papers citing it
Background
82% of classified citations
hub tools
citation-role summary
background 37
method 3
other 2
baseline 1
dataset 1
citation-polarity summary
authors
co-cited works
representative citing papers
2G2T enables constant-size, statistically sound outsourcing of MSM with verification up to 300x faster than local computation and error probability at most 1/q.
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 dependent linear type theory is constructed by embedding linear logic into dependent type theory, yielding multiplicities that depend on variables, supporting W-types, with semantics in indexed Categories with Families and an Agda implementation.
HDL defines dynamic theories with lifting and combination operations, proves soundness and relative completeness in Isabelle, and demonstrates the approach on a Java controller steering a differential dynamic logic plant model.
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.
A differentiable neural framework for learning state- and time-dependent parameters of finite-state mean field games from population trajectories via implicit differentiation.
MeiBRD meta-learns a graph-neural residual deformation function to correct linear biomechanical predictions for intraoperative liver registration from sparse context samples.
SpikeTAD proposes the first SNN-based end-to-end TAD model, reporting 67.2% mAP on THUMOS14 and 37.42% on ActivityNet-1.3 with extremely low power consumption.
For lopsided trees with t2 >= 2 t1, Burr's bound has a gap of order max(t1^2/t2, sqrt(t1)); for t2 >= 500 t1 the bound is tight if Delta(T) <= t2 - t1 but off by Omega(log t2) otherwise.
Recurrence of x-periodic rogue waves in Q1D MNLS shows model-dependent fission/fusion and is described analytically by finite gap perturbation theory of NLS AWs with good numerical agreement.
The paper proposes the ontological continuum as a new theoretical construct defined by two orthogonal distinctions to provide a vocabulary for describing, comparing, navigating, and transforming knowledge graphs across modeling practices.
An O(n^5) exact algorithm for broadcast domination obtained by reducing the path-case to O(n^3) via a single DAG on oriented broadcast balls.
Polyconvexity implies true-stress-true-strain monotonicity in incompressible isotropic hyperelasticity, which is enforced in four PANN architectures that show varying extrapolation behavior on experimental data.
Event-B Agent is an LLM agent that synthesizes, refines, and repairs Event-B formal models from natural language requirements via iterative verification feedback loops.
CDCL SAT solvers compute small proofdoors on linearly scaling BMC families but large non-absorbed ones on exponential families, as shown by empirical measurements on a 76k+ instance benchmark where prior parameters fail to discriminate regimes.
Proves that ℓ_p norm minimization yields p-independent Hausdorff convergence rate O(k^{2/(1-q)}) in convex vector optimization via Euclidean intermediary and norm equivalence.
A compositional algebraic decision diagram algorithm quantifies sensitivity in decision tree ensembles with certified error and confidence bounds, outperforming model counters on benchmarks.
PAFP is FPT by BFS-width plus backward arcs in the union digraph and polynomial-time solvable via 2-SAT for DAGs of exact-length width 2, with matching NP-hardness for width 3.
FLiD is a field-localized forgery detection method for identity documents that outperforms full-document baselines and general detectors with significantly fewer parameters.
LLM2Ltac mines symbolic tactics from 11,725 Coq theorems using LLMs and integrates them into CoqHammer, improving proof rates by 23.87% on 6,199 theorems from four large verification projects.
A source-level interaction concept for interactive program verification, prototyped in KeY, improves user understanding of proof states and defect detection according to a user study.
SCHORTY uses a tilted sensor in a single camera to map pixel position to range via the Scheimpflug principle, demonstrated experimentally with event cameras for passive UAV ranging up to 1.1 km.
Text-guided class-agnostic counting models exhibit significant weaknesses in grounding textual prompts to visual objects, as demonstrated by new negative-label and distractor tests on a multi-category dataset.
citing papers explorer
-
Learning Spectral and Polarimetric Clues for One-to-Multimodal Novel View Synthesis
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.
-
MeiBRD: Meta-Learning Intraoperative Biomechanical Residual Deformation
MeiBRD meta-learns a graph-neural residual deformation function to correct linear biomechanical predictions for intraoperative liver registration from sparse context samples.
-
SpikeTAD: Spiking Neural Networks for End-to-End Temporal Action Detection
SpikeTAD proposes the first SNN-based end-to-end TAD model, reporting 67.2% mAP on THUMOS14 and 37.42% on ActivityNet-1.3 with extremely low power consumption.
-
Field-Localized Forgery Detection for Digital Identity Documents
FLiD is a field-localized forgery detection method for identity documents that outperforms full-document baselines and general detectors with significantly fewer parameters.
-
Does it Really Count? Assessing Semantic Grounding in Text-Guided Class-Agnostic Counting
Text-guided class-agnostic counting models exhibit significant weaknesses in grounding textual prompts to visual objects, as demonstrated by new negative-label and distractor tests on a multi-category dataset.
-
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.
-
Appear2Meaning: A Cross-Cultural Benchmark for Structured Cultural Metadata Inference from Images
A new cross-cultural benchmark shows vision-language models infer structured cultural metadata from images inconsistently, with fragmented signals and large performance gaps across regions and metadata types.
-
SOMA: From Surface Observations to Muscle Anatomy
SOMA recovers spatio-temporal muscle behavior from multi-view RGB surface data and introduces the SKIM soft-tissue deformation dataset as the first such method from RGB observations.
-
Unlocking UML Class Diagram Understanding in Vision Language Models
A new UML class diagram VQA benchmark and 16k dataset enable LoRA fine-tuning to outperform Qwen 3.5 27B.
-
Clinically Aligned Geometry Constraints for Robust IVUS Vessel Boundary Segmentation
GeoCat combines dual Cartesian-polar encoders with a geometry consistency loss to improve both segmentation overlap and clinical geometry accuracy on a 12k-frame IVUS dataset.
-
CHROMA: Detecting AI-Generated Images through Inter-Channel Color-Space Correlations
CHROMA augments RGB images with inter-channel correlation maps from multiple color spaces and trains a fixed CNN to detect AI-generated images, achieving competitive performance with limited multi-generator data.
-
Clustering Guided Domain-Specific Pretrained Foundation Model for Very High-Resolution Arctic Remote Sensing
Affinity-propagation clustering of Arctic VHSR imagery enables MAE pretraining of a ViT-Large encoder that outperforms ImageNet and Prithvi-EO-2.0 baselines by 5-15 percentage points in mean F1 on four downstream Arctic detection and segmentation tasks.
-
Generator-Refiner-Examiner: A Tri-Module Data Augmentation Framework for 3D Human Avatar Learning from Monocular Videos
TrioMan is a tri-module data augmentation framework using a Generator for pose/camera perturbations, a Refiner with one-step diffusion, and an Examiner with dual-branch attention to improve 3D avatar learning from monocular videos, claiming better results than prior methods on two benchmarks.
-
Vision Transformer-Conditioned UNet for Domain-Adaptive Semantic Segmentation
ViTC-UNet adapts frozen ViT representations to biomedical semantic segmentation by conditioning a UNet via learnable tokens and two-way attention decoding.
-
Supersampling Stable Diffusion and Beyond: A Seamless, Training-Free Approach for Scaling Neural Networks Using Common Interpolation Methods
Kernel interpolation with a constant multiplier scales convolution and fully-connected layers in neural networks to higher resolutions or dimensions without training, producing competitive results on Stable Diffusion and other models.
-
Edge-Constrained UAV Small-Object Detection with P2 Enhancement and Quantum-Inspired Lightweight Structure Search
Adding a P2 branch to YOLOX-Nano raises small-object AP by 31.10% on VisDrone; QIEA screens structures balancing accuracy, FLOPs, latency, memory and recall.
-
FLIM Networks with Bag of Feature Points
FLIM-BoFP replaces per-block patch clustering in FLIM networks with a single input-level clustering step that creates a bag of feature points used to define filters across all encoder blocks, yielding faster training for parasite detection in optical microscopy.