Negative / Null Result ReportOpen accessComputer Science
Jin Zeng, Yang Liu, Gene Cheung et al. · 2022 · arXiv
A graph convolutional network (GCN) employs a graph filtering kernel tailored for data with irregular structures. However, simply stacking more GCN layers does not improve performance; instead, the output converges to an uninformative low-dimensional subspace, where the convergence rate is characterized by the graph spectrum -- this is the known over-smoothing problem in GCN. In this paper, we propose a sparse graph learning algorithm incorporating a new spectrum prior to compute a graph topology that circumvents over-smoothing while preserving pairwise correlations inherent in data. Specifica
Negative / Null Result ReportOpen accessComputer Science
Wentao Zhao, Dalin Zhou, Xinguo Qiu et al. · 2020 · arXiv
Graph neural networks (GNNs) have been investigated for potential applicability in multiple fields that employ graph data. However, there are no standard training settings to ensure fair comparisons among new methods, including different model architectures and data augmentation techniques. We introduce a standard, reproducible benchmark to which the same training settings can be applied for node classification. For this benchmark, we constructed 9 datasets, including both small- and medium-scale datasets from different fields, and 7 different models. We design a k-fold model assessment strate
Negative / Null Result ReportOpen accessComputer Science
Charles Jin, Melinda Sun, Martin Rinard · 2021 · arXiv
We propose a novel clustering mechanism based on an incompatibility property between subsets of data that emerges during model training. This mechanism partitions the dataset into subsets that generalize only to themselves, i.e., training on one subset does not improve performance on the other subsets. Leveraging the interaction between the dataset and the training process, our clustering mechanism partitions datasets into clusters that are defined by--and therefore meaningful to--the objective of the training process. We apply our clustering mechanism to defend against data poisoning attacks,
Negative / Null Result ReportOpen accessPhysics
Fabien Robineau, Frédéric Boy, Jean-Pierre Orliaguet et al. · 2006 · arXiv
Performing minimal-invasive surgical punctures require guiding a needle toward an intracorporeal clinically-defined target. As this technique does not involve cutting the body open, a visualization system is employed to provide the surgeon with indirect visual spatial information about the intracorporeal positions of the tool. One may consider that such systems reduce the ergonomics of the situation as they generate a decorrelation between the actual movement of the needle and the displayed information about this movement. The present study aims at assessing the guidance of an intracorporeal p
Negative / Null Result ReportOpen accessComputer Science
Alexia Jolicoeur-Martineau · 2019 · arXiv
This paper provides a more rigorous look at Relativistic Generative Adversarial Networks (RGANs). We prove that the objective function of the discriminator is a statistical divergence for any concave function $f$ with minimal properties ($f(0)=0$, $f'(0) \neq 0$, $\sup_x f(x)>0$). We also devise a few variants of relativistic $f$-divergences. Wasserstein GAN was originally justified by the idea that the Wasserstein distance (WD) is most sensible because it is weak (i.e., it induces a weak topology). We show that the WD is weaker than $f$-divergences which are weaker than relativistic $f$-diver
Negative / Null Result ReportOpen accessEngineering
Neil Irwin Bernardo, Jingge Zhu, Jamie Evans · 2021 · arXiv
We analyze the symbol error probability (SEP) of $M$-ary pulse amplitude modulation ($M$-PAM) receivers equipped with optimal low-resolution quantizers. We first show that the optimum detector can be reduced to a simple decision rule. Using this simplification, an exact SEP expression for quantized $M$-PAM receivers is obtained when Nakagami-$m$ fading channel is considered. The derived expression enables the optimization of the quantizer and/or constellation under the minimum SEP criterion. Our analysis of optimal quantization for equidistant $M$-PAM receiver reveals the existence of error fl
Negative / Null Result ReportOpen accessComputer Science
Zachary Baker, Yuxiao Li · 2025 · arXiv
Sparse Autoencoders (SAEs) have emerged as a promising approach for interpreting neural network representations by learning sparse, human-interpretable features from dense activations. We investigate whether incorporating variational methods into SAE architectures can improve feature organization and interpretability. We introduce the Variational Sparse Autoencoder (vSAE), which replaces deterministic ReLU gating with stochastic sampling from learned Gaussian posteriors and incorporates KL divergence regularization toward a standard normal prior. Our hypothesis is that this probabilistic sampl
Negative / Null Result ReportOpen accessMathematics
Ben Adcock, Gregor Maier, Rahul Parhi · 2025 · arXiv
We develop a minimax theory for operator learning, where the goal is to estimate an unknown operator between separable Hilbert spaces from finitely many noisy input-output samples. For uniformly bounded Lipschitz operators, we prove information-theoretic lower bounds together with matching or near-matching upper bounds, covering both fixed and random designs under Hilbert-valued Gaussian noise and Gaussian white noise errors. The rates are controlled by the spectrum of the covariance operator of the measure that defines the error metric. Our setup is very general and allows for measures with u
Negative / Null Result ReportOpen accessComputer Science
Ali Subhan, Ashir Raza · 2026 · arXiv
DragDiffusion is a diffusion-based method for interactive point-based image editing that enables users to manipulate images by directly dragging selected points. The method claims that accurate spatial control can be achieved by optimizing a single diffusion latent at an intermediate timestep, together with identity-preserving fine-tuning and spatial regularization. This work presents a reproducibility study of DragDiffusion using the authors' released implementation and the DragBench benchmark. We reproduce the main ablation studies on diffusion timestep selection, LoRA-based fine-tuning, mas
Replication FailureOpen accessComputer Science
Zhichao Fan, Yanhang Li, Zexin Zhuang · 2026 · arXiv
Forced chain-of-thought (CoT) is widely assumed to make vision-language models more reliable on video question answering. We propose a small three-probe evaluation recipe to test that assumption: paired accuracy across direct, CoT, answer-first, and no-video conditions; a counterfactual video-swap diagnostic over the CoT chains; and a four-rung visual-degradation ladder. Each probe is reported under both a strict and a permissive regex scorer, with multiplicity correction over a manuscript-declared primary family. Applied to Qwen2.5-VL on Video-MME subsets, the recipe returns a two-part findin
Replication FailureOpen accessComputer Science
Xiqi Hao, Zengqing Wu, Yu-Xuan Qiu et al. · 2026 · arXiv
Multi-agent debate (MAD) is a promising strategy for improving LLM reasoning, but when agents converge on a shared answer, it is unclear whether that convergence reflects genuine deliberation or social compliance. We show that the conventional answer flip rate conflates three distinct mechanisms: spontaneous instability, stance-induced conformity, and reasoning-induced persuasion. Our three-source decomposition framework isolates each through controlled counterfactual conditions. In the primary MMLU-Pro setting, 37% of agent-question observations change under self-reflection alone, while robus
Replication FailureOpen accessComputer Science
J. Jon Ryu, Pavan Yeddanapudi, Xiangxiang Xu et al. · 2025 · arXiv
The InfoNCE objective, originally introduced for contrastive representation learning, has become a popular choice for mutual information (MI) estimation, despite its indirect connection to MI. In this paper, we demonstrate why InfoNCE should not be regarded as a valid MI estimator, and we introduce a simple modification, which we refer to as InfoNCE-anchor, for accurate MI estimation. Our modification introduces an auxiliary anchor class, enabling consistent density ratio estimation and yielding a plug-in MI estimator with significantly reduced bias. Beyond this, we generalize our framework us
Replication FailureOpen accessMathematics
Natesh Pillai, Aaron Smith · 2026 · arXiv
In Diaconis and Saloff-Coste (1996), the authors introduced the simple ``transvection" walk on $\mathrm{GL}_n(\mathbb F_2)$: at each step, choose two distinct rows and add one to the other. In Ben-Hamou (2025), the author recently proved that this walk has mixing time $O(n^2\log n)$. Inspired by applications in cryptography (see Sotiraki (2016)), Ben-Hamou and Peres (2018) conjectured that the first $k$ columns of this walk mixed in $O(nk \log(n))$ steps. Our main result is a proof of this conjecture uniformly in $n$ and $k.$ Our proof is based on a local-to-global entropy estimate, in the spi
Negative / Null Result ReportOpen accessComputer Science
Paul Röttger, Janet B. Pierrehumbert · 2021 · arXiv
Language use differs between domains and even within a domain, language use changes over time. For pre-trained language models like BERT, domain adaptation through continued pre-training has been shown to improve performance on in-domain downstream tasks. In this article, we investigate whether temporal adaptation can bring additional benefits. For this purpose, we introduce a corpus of social media comments sampled over three years. It contains unlabelled data for adaptation and evaluation on an upstream masked language modelling task as well as labelled data for fine-tuning and evaluation on
Negative / Null Result ReportOpen accessComputer Science
Mireille El-Assal, Pierre Tirilly, Ioan Marius Bilasco · 2023 · arXiv
Video analysis is a computer vision task that is useful for many applications like surveillance, human-machine interaction, and autonomous vehicles. Deep Convolutional Neural Networks (CNNs) are currently the state-of-the-art methods for video analysis. However they have high computational costs, and need a large amount of labeled data for training. In this paper, we use Convolutional Spiking Neural Networks (CSNNs) trained with the unsupervised Spike Timing-Dependent Plasticity (STDP) learning rule for action classification. These networks represent the information using asynchronous low-ener
Negative / Null Result ReportOpen accessComputer Science
Óscar García-Sierra, Ana Fernández-Pampillón Cesteros, Miguel Ortega-Martín · 2024 · arXiv
Subword tokenization algorithms used by Large Language Models are significantly more efficient and can independently build the necessary vocabulary of words and subwords without human intervention. However, those subwords do not always align with real morphemes, potentially impacting the models' performance, though it remains uncertain when this might occur. In previous research, we proposed a method to assess the morphological quality of vocabularies, focusing on the overlap between these vocabularies and the morphemes of a given language. Our evaluation method was built on three quality meas
Negative / Null Result ReportOpen accessNutrition. Foods and food supply
Kaijun Wang, Kaijun Wang, Jie Ma et al. · 2022 · Frontiers in Nutrition
Artemisia argyi leaf is a well-known species in traditional Chinese medicine, and its essential oil (AAEO) has been identified to exert various physiological activities. The aim of this study was to investigate the effects of AAEO on lipid metabolism and the potential microbial role in high-fat diet (HFD)-fed mice. A total of 50 male mice were assigned to five groups for feeding with a control diet (Con), a high-fat diet (HFD), and the HFD plus the low (LEO), medium (MEO), and high (HEO) doses of AAEO. The results demonstrated that dietary HFD markedly increased the body weight gain compared w
View details →DOI: 10.3389/fnut.2022.1024722 Negative / Null Result ReportOpen accessMedicine
Julie Labrosse, Marie Osdoit, Anne-Sophie Hamy et al. · 2020 · PLoS ONE
Although identified to be at a higher risk of relapse, no consensus exists on the treatment of breast cancer (BC) patients with no pathological complete response after neoadjuvant chemotherapy (NAC). The benefit of adjuvant chemotherapy (ADJ) in this context has scarcely been studied. We evaluated the benefit of administrating adjuvant chemotherapy in a real life cohort of BC patients with invasive residual disease after NAC. 1199 female BC patients with T1-3NxM0 invasive tumors receiving NAC at Institut Curie from 2002 to 2012 were included in the analysis. 1061 had been treated by NAC only,
View details →DOI: 10.1371/journal.pone.0234173 Negative / Null Result ReportOpen accessOrthopedic surgery
Yakup Ekinci, Mithat Oner, Ibrahim Karaman et al. · 2020 · Acta Orthopaedica et Traumatologica Turcica
Abstract Objective: The aim of this study was to compare simultaneous bilateral total knee arthroplasty (BTKA) and unilateral total knee arthroplasty (UTKA) in terms of morbidity, clinical and radiological findings and quality of life. Methods: The study included 48 simultaneous BTKAs (46 females, 2 males; mean age: 64.00±8.31 years) and 53 UTKAs (46 females, 7 males; mean age: 64.40±7.45 years) performed between November 2007 and June 2012. Groups were compared with respect to comorbidity, complications, blood transfusion, hospital stay, clinical and radiological (American Knee Society Score)
View details →DOI: 10.3944/aott.2014.3226 Negative / Null Result ReportOpen accessArctic medicine. Tropical medicine
Yuelin Xiong, Huatang Luo, Hao Wang et al. · 2026 · PLoS Neglected Tropical Diseases
Background Schistosomiasis, a neglected tropical disease (NTD), remains a public health concern in China. Health education is a fundamental intervention for its control. Even in transmission-interrupted areas like Wuhan, sustained awareness is crucial. However, recent literature on school-based interventions evaluating knowledge, attitudes and practices (KAP) among children in such areas is limited. Objective This study aimed to evaluate and compare the effectiveness of peer-led education versus routine health education in improving schistosomiasis-related KAP among elementary students in an u
View details →DOI: 10.1371/journal.pntd.0013857 Negative / Null Result ReportOpen accessOphthalmology
Jia-Wen Li, Yi-Jian Li, Xi-Su Hu et al. · 2020 · International Journal of Ophthalmology
AIM: To assess the biosafety of a poly(acrylamide-co-sodium acrylate) hydrogel (PAH) as a 3D-printed intraocular lens (IOL) material. METHODS: The biosafety of PAH was first evaluated in vitro using human lens epithelial cells (LECs) and the ARPE19 cell line, and a cell counting kit-8 (CCK-8) assay was performed to investigate alterations in cell proliferation. A thin film of PAH and a conventional IOL were intraocularly implanted into the eyes of New Zealand white rabbits respectively, and a sham surgery served as control group. The anterior segment photographs, intraocular pressure (IOP), bl
View details →DOI: 10.18240/ijo.2020.10.03 Negative / Null Result ReportOpen accessDentistry
Fatemeh Maleknejad, Saied Mostafa Moazzami, Elham Baradaran Nasseri et al. · 2011 · Journal of Dental Research, Dental Clinics, Dental Prospects
Background and aims. Adhesive permeability is hindered by application of an additional layer of hydrophobic resin, which increases its concentration within the hydrophilic layer, reduces its affinity to water, and enhances its physical properties. The aim of the present study was to evaluate the effect of a hydrophobic layer on the microleakage of class V composite restorations using different adhesives. Materials and methods. The adhesives including total-etch Scotchbond MP and Single Bond, and the self-etch Clearfil SE Bond and Clearfil S3 Bond were applied to 80 class V cavities in vitro on
Negative / Null Result ReportOpen accessMedicine
Yong Xue, Ai Zhao, Li Cai et al. · 2015 · PLoS ONE
OBJECTIVE:To explore the associations between picky eating behaviour and pre-schoolers' growth and development. Corresponding potential mechanisms, such as nutrient and food subgroup intake, as well as micronutrients in the blood, will be considered. METHODS:Picky eating behaviour was present if it was reported by parents. From various areas of China, 937 healthy children of 3-7 years old were recruited using a multi-stage stratified cluster sampling method. Children and their mothers' socio-demographic information and children's anthropometry, intelligence, blood samples, one 24-hour dietary
View details →DOI: 10.1371/journal.pone.0123664 Negative / Null Result ReportOpen accessInternal medicine
Tri Ratnaningsih, Novi W Sukirto, Arum Tri Wahyuningsih · 2020 · Acta Medica Indonesiana
Background: iron deficiency in pulmonary tuberculosis (TB) patients may weaken their immune system, causing difficulty in overcoming the infection. Accurate diagnosis of iron deficiency anemia (IDA) in pulmonary TB patients is essential. In order to prove the iron deficient state, diagnosis should focus on inflammatory factors, which could highly affect the outcome of iron status, such as measurement of serum ferritin (SF). Soluble Transferrin Receptor (sTfR) is the best parameter to diagnose iron deficiency in the inflammatory condition. This study aimed to understand the role of sTfR to iden
Negative / Null Result ReportOpen accessDiseases of the musculoskeletal system
Mohamed A. Mortada, Haidy E. Zidan, Noha A. Abdelsalam et al. · 2020 · Egyptian Rheumatology and Rehabilitation
Abstract Background Impaired deoxyribonucleic acid (DNA) repair may induce an autoimmune response in susceptible individuals. The association between DNA repair gene XRCC1 Arg399Gln gene polymorphism and susceptibility of systemic lupus erythematosus (SLE) is controversial. This study aimed to detect the association of XRCC1 Arg399Gln gene polymorphism with SLE and its clinical manifestations in the Egyptian population. Results A significant association was found between weight loss and genotype GG (P = 0.048); however, genotype AG was significantly associated with high serum creatinine and lo
View details →DOI: 10.1186/s43166-020-00009-2 Negative / Null Result ReportOpen accessImmunologic diseases. Allergy
Hiroyuki Kawahara, Satoshi Hara, Noriko Iwaki et al. · 2025 · Immunological Medicine
TAFRO syndrome is a systemic inflammatory disorder of unknown etiology, and its diagnosis requires the exclusion of autoimmune diseases. A 42-year-old Japanese woman presented with TAFRO syndrome-like manifestations, but had undiagnosed limited-cutaneous systemic sclerosis preventing a definitive diagnosis of TAFRO syndrome. However, her clinical course and pathological findings, including renal glomerular microangiopathy, were consistent with TAFRO syndrome. We performed a systematic review of the literature to evaluate how autoimmunity affects the clinical characteristics of TAFRO syndrome/i
View details →DOI: 10.1080/25785826.2025.2481675 Negative / Null Result ReportOpen accessMedicine
Amr Almaz Abdel-aziem, Amira Hussin Draz · 2013 · Medicine Science
The purpose of this study was to examine the effect of chronic ankle instability (CAI) on eccentric peak torque of the ankle musculature. The study included fifty participants divided into two groups; the first group consisted of twenty five subjects suffering from CAI, the second group consisted of twenty five subjects with no prior history of ankle injury as control group. Eccentric peak torque of invertors, evertors, plantar-flexors and dorsiflexors was recorded for each participant at angular velocities 30 and 90o/sec by using (Biodex Medical Systems, Inc, Shirley, NY, USA). Analysis of va
View details →DOI: 10.5455/medscience.2013.02.8048 Negative / Null Result ReportOpen accessMedicine (General)
A.G. Elsaid, N.M. Faheem · Brazilian Journal of Medical and Biological Research
Continuous industrial productivity and modern societies have resulted in excess artificial light. The altered circadian rhythm causes many diseases. During intrauterine life, the mother's maternal melatonin rhythm has a major role in influencing organ development. The aim of this study was to investigate the effect of maternal exposure to constant light on the structure and ultrastructure of neonatal skin. Twenty pregnant New Zealand rabbits were divided into two groups (n=10 each): control group (12-h light/dark) and constant light group (24-h light). Plasma maternal melatonin and corticoster
View details →DOI: 10.1590/1414-431x202010722 Negative / Null Result ReportOpen accessOrthopedic surgery
Ahmet Salduz, Fatih Dikici, Önder Ismet Kılıçoğlu et al. · 2017 · Journal of Orthopaedic Surgery
Purpose: The aim of our study is to investigate the bone ongrowth of two different alternative surfaces and the effect of nonsteroidal anti-inflammatory drugs (NSAIDs) on osseointegration. Methods: Hips of 40 New Zealand white rabbits were operated bilaterally. Hydroxyapatite (HA)-coated titanium rods were implanted into the right femur, and grit-blasted titanium rods were implanted into the left femur. They were divided into three groups. At the end of 8 weeks, both femora of the rabbits were removed and investigated biomechanically and histologically. Results: HA-coated implants had a signif
View details →DOI: 10.1177/2309499016684410 Negative / Null Result ReportOpen accessMedicine
XUE Bing, XI Hualei, YAO Lihong, XU Wanqiu, XU Xiaohang, LIN Song, PIAO Guiyan, WANG Xiumei · 2025 · 口腔疾病防治
Objective To investigate the effect of neurite outgrowth inhibitor extracellular peptide residues 1-40 (NEP1-40) combined with poly (lactic-co-glycolic acid) (PLGA) and gelatin electrospun fiber membrane on facial nerve repair in rats. Methods According to the principle of random grouping, 108 male SD rats were divided into four groups ( n = 27 in each group, approved by the ethics committee), namely, the sham group, control group, PLGA group, and NEP1-40 + PLGA group. A facial nerve fracture model was established for all of the groups except for the sham group. The control group received no f
View details →DOI: 10.12016/j.issn.2096-1456.202440237 