Negative / Null Result ReportOpen accessComputer Science
Andrei Liviu Nicolicioiu, Mohammad Pezeshki, Aaron Courville · 2026 · arXiv
On-policy self-distillation achieves strong pass@1 accuracy by using a single model as both teacher and student, with the teacher conditioned on a correct demonstration to provide dense token-level feedback. We show that this could come at a hidden cost: rollout diversity decreases and pass@k curves flatten (i.e., generating more rollouts fails to improve accuracy). We trace this to compounding biases in the design of self-distillation with sampled demonstrations. The teacher scores each student rollout while conditioned on a sampled correct rollout, channeling its feedback through the model's
Negative / Null Result ReportOpen accessGeriatrics
Lars Donath, Sebastian Ludyga, Daniel Hammes et al. · 2017 · BMC Geriatrics
Abstract Background Aging is accompanied by a decline of executive function. Aerobic exercise training induces moderate improvements of cognitive domains (i.e., attention, processing, executive function, memory) in seniors. Most conclusive data are obtained from studies with dementia or cognitive impairment. Confident detection of exercise training effects requires adequate between-day reliability and low day-to-day variability obtained from acute studies, respectively. These absolute and relative reliability measures have not yet been examined for a single aerobic training session in seniors.
View details →DOI: 10.1186/s12877-017-0634-x Negative / Null Result ReportOpen accessMedicine
Niroshica Mohanathas, Lianna Montanari, Rachel Downey et al. · 2026 · PLoS ONE
Background Multitasking, such as listening while balancing, relies on integrated processing in the sensory, cognitive, and motor systems; systems that often decline with age. Hearing loss is linked to increased risks of both falls and cognitive decline. Improving cognitive processing through executive function (EF) training may support balance, especially in older adults with hearing loss. This randomized controlled study conducted across age groups and hearing abilities, examined the effects of a 12-week EF training program on postural outcomes (center of pressure (COP)) using an auditory-cog
View details →DOI: 10.1371/journal.pone.0331276 Negative / Null Result ReportMedicine
Pan, Luan, Zhang et al. · 2026 · PloS one
Scale-Invariant Feature Transform (SIFT) features are widely used in target recognition and tracking. This study aimed to exploit the robust performance of the SIFT algorithm to accurately calculate tumor tissue displacement during…
View details →DOI: 10.1371/journal.pone.0330578 Negative / Null Result ReportOpen accessComputer Science
Lennart Luettgau, Vanessa Cheung, Magda Dubois et al. · 2025 · arXiv
People increasingly seek personal advice from large language models (LLMs), yet whether humans follow their advice, and its consequences for their well-being, remains unknown. In a longitudinal randomised controlled trial with a representative UK sample (N = 6,474), we found that up to 79% of participants who had a 20-minute discussion with one of three AI chatbots (GPT-4o, LLama-3.3-70B, Gemini 3 Pro) about health, careers or relationships subsequently reported following its advice. Advice-following remained above 60% even for high-stakes recommendations, suggesting that users only weakly cal
Negative / Null Result ReportOpen accessComputer Science
Ruth Cohen, Lu Feng, Ayala Bloch et al. · 2026 · arXiv
While natural-language explanations from large language models (LLMs) are widely adopted to improve transparency and trust, their impact on objective human-AI team performance remains poorly understood. We identify a Persuasion Paradox: fluent explanations systematically increase user confidence and reliance on AI without reliably improving, and in some cases undermining, task accuracy. Across three controlled human-subject studies spanning abstract visual reasoning (RAVEN matrices) and deductive logical reasoning (LSAT problems), we disentangle the effects of AI predictions and explanations u
Negative / Null Result ReportOpen accessComputer Science
Amanpreet Singh, Mike D'Arcy, Arman Cohan et al. · 2022 · arXiv
Learned representations of scientific documents can serve as valuable input features for downstream tasks without further fine-tuning. However, existing benchmarks for evaluating these representations fail to capture the diversity of relevant tasks. In response, we introduce SciRepEval, the first comprehensive benchmark for training and evaluating scientific document representations. It includes 24 challenging and realistic tasks, 8 of which are new, across four formats: classification, regression, ranking and search. We then use this benchmark to study and improve the generalization ability o
Negative / Null Result ReportMedicine
Zolfaghari, Nouri Habashi, Aghazadeh · 2026 · Veterinary research forum : an international quarterly journal
Polycystic ovary syndrome (PCOS) is a complex endocrine disorder. Exercise is one of the interventions that leads to PCOS improvement. This study aimed to investigate the potential of different durations of high-intensity interval training…
View details →DOI: 10.30466/vrf.2025.2050200.4623 Negative / Null Result ReportOpen accessComputer Science
Chunliang Li, Tianze Cao, Sanyuan Zhao · 2026 · arXiv
Visual Autoregressive (VAR) modeling inefficiently applies a fixed computational depth to each position when generating high-resolution images. While existing methods accelerate inference by pruning tokens using frequency maps, their binary hard-pruning approach is fundamentally limited and fails to improve quality even with better frequency estimation. Observing that VAR models possess significant depth redundancy, we propose a paradigm shift from pruning entire tokens to adaptively allocating per-token computational depth. To this end, we introduce DepthVAR, a training-free framework that dy
Negative / Null Result ReportOpen accessVeterinary medicine
Pei-Chih Lee, Yuqing Yang, Kaywalee Chatdarong et al. · 2026 · Frontiers in Veterinary Science
IntroductionDry-preservation of gametes offers a promising alternative to cryobanking by potentially storing cells at ambient temperatures; however, successful preservation of whole oocytes remains limited by dehydration-induced cellular damages. This study evaluated whether intracellular delivery of trehalose via a membrane-permeable derivative, trehalose hexaacetate (6-O-Ac-Tre), could enhance dehydration tolerance of domestic cat cumulus–oocyte complexes (COCs), a relevant model for mammalian fertility preservation.MethodsCat COCs were incubated with 0, 3, 10, or 30 mM of 6-O-Ac-Tre to asse
View details →DOI: 10.3389/fvets.2026.1805930 Negative / Null Result ReportOpen accessImmunologic diseases. Allergy
Yuka Nadai, Yuka Nadai, Kathrin Held et al. · 2019 · Frontiers in Immunology
Background: A better understanding of the parameters influencing vaccine-induced IgG recognition of individual antigenic regions and their variants within the HIV Envelope protein (Env) can help to improve design of preventive HIV vaccines.Methods: Env-specific IgG responses were mapped in samples of the UKHVC003 Standard Group (UK003SG, n = 11 from UK) and TaMoVac01 (TMV01, n = 17 from Tanzania) HIV vaccine trials. Both trials consisted of three immunizations with DNA, followed by two boosts with recombinant Modified Vaccinia Virus Ankara (MVA), either mediating secretion of gp120 (UK003SG) o
View details →DOI: 10.3389/fimmu.2019.00717 Negative / Null Result ReportOpen accessBotany
R. mawar, S. Lodha · 2002 · Phytopathologia Mediterranea
The combined effect of Brassica amendments (mustard oil–cake or mustard residue at 2.5 tons ha-1) and summer irrigation was tested for survival of Macrophomina phaseolina and Fusarium oxysporum f. sp. cumini and on the severity of dry root rot on clusterbean in the rainy season and on wilt of cumin in the subsequent winter season in the same field. Seed coating with a Bacillus sp., an antagonist bacterium against M. phaseolina, was also integrated with pod residues to improve control of dry root rot. The soil temperature of amended soil after one summer irrigation in June ranged from 38–44°C a
View details →DOI: 10.14601/phytopathol_mediterr-1649 Negative / Null Result ReportOpen accessComputer Science
Adam Byerly, Daniel Khashabi · 2024 · arXiv
Self-consistency (SC) improves the performance of large language models (LLMs) across various tasks and domains that involve short content. However, does this support its effectiveness for long-context problems? We challenge the assumption that SC's benefits generalize to long-context settings, where LLMs often struggle with position bias, the systematic over-reliance on specific context regions-which hinders their ability to utilize information effectively from all parts of their context. Through comprehensive experimentation with varying state-of-the-art models, tasks, and SC formulations, w
Negative / Null Result ReportOpen accessMedical emergencies. Critical care. Intensive care. First aid
Christopher J. Kirwan, Barbara J. Philips, Iain A. M. MacPhee · 2013 · Critical Care Research and Practice
Introduction. RIFLE and AKIN provide a standardised classification of acute kidney injury (AKI), but their categorical rather than continuous nature restricts their use to a research tool. A more accurate real-time description of renal function in AKI is needed, and some published data suggest that equations based on serum creatinine that estimate glomerular filtration rate (eGFR) can provide this. In addition, incorporating serum cystatin C concentration into estimates of GFR may improve their accuracy, but no eGFR equations are validated in critically ill patients with AKI. Aim. This study t
View details →DOI: 10.1155/2013/406075 Negative / Null Result ReportOpen accessComputer Science
Aaron B. Wagner, Nirmal V. Shende, Yücel Altuğ · 2019 · arXiv
We introduce a novel mechanism, called timid/bold coding, by which feedback can be used to improve coding performance. For a certain class of DMCs, called compound-dispersion channels, we show that timid/bold coding allows for an improved second-order coding rate compared with coding without feedback. For DMCs that are not compound dispersion, we show that feedback does not improve the second-order coding rate. Thus we completely determine the class of DMCs for which feedback improves the second-order coding rate. An upper bound on the second-order coding rate is provided for compound-dispersi
Negative / Null Result ReportOpen accessSpecialties of internal medicine
Colleen M. Badke, Matthew L. Friedman, Z. Leah Harris et al. · 2020 · Resuscitation Plus
Aim: To determine if an untrained cardiopulmonary resuscitation (CPR) Coach, with no access to real-time CPR feedback technology, improves CPR quality. Methods: This was a prospective randomized pilot study at a tertiary care children's hospital that aimed to integrate an untrained CPR Coach into resuscitation teams during simulated pediatric cardiac arrest. Simulation events were randomized to two arms: control (no CPR Coach) or intervention (CPR Coach). Simulations were run by pediatric intensive care unit (PICU) providers and video recorded. Scenarios focused on full cardiopulmonary arrest;
View details →DOI: 10.1016/j.resplu.2020.100035 Negative / Null Result ReportOpen accessAgricultural and Biological Sciences
Denis Boyer, Peter D. Walsh · 2010 · arXiv
Thanks to recent technological advances, it is now possible to track with an unprecedented precision and for long periods of time the movement patterns of many living organisms in their habitat. The increasing amount of data available on single trajectories offers the possibility of understanding how animals move and of testing basic movement models. Random walks have long represented the main description for micro-organisms and have also been useful to understand the foraging behaviour of large animals. Nevertheless, most vertebrates, in particular humans and other primates, rely on sophistic
Negative / Null Result ReportOpen accessPsychiatry
Jiaojiao Zhou, Xu Chen, Le Xiao et al. · 2021 · Frontiers in Psychiatry
Background: Insomnia is a common clinical manifestation in patients with depression. Insomnia is not only a depression symptom but also an independent risk factor for recurrence. Cordyceps militaris (C. militaris) is thought to have the potential to treat insomnia. This study aimed to examine the efficacy and safety of duloxetine with C. militaris in improving sleep symptoms in patients with depression.Methods: This study was a single-center, randomized, double-blind, placebo-controlled study that recruited outpatients admitted to Beijing Anding hospital from January 2018 to January 2019. Majo
View details →DOI: 10.3389/fpsyt.2021.754921 Negative / Null Result ReportOpen accessInternal medicine
Xiangyan Liu, Tian Shen, Zhuoyi Wang et al. · 2014 · Journal of Infection in Developing Countries
Hepatitis E virus (HEV) infection in most individuals is known as a self-limiting, acute, icteric hepatitis, but evidence shows HEV is responsible for choric hepatitis and rapid progressed liver cirrhosis in immuno-compromised patients. We present the case of a patient whose diagnosis of acute graft failure was due to a HEV infection 7 years after his first liver transplantation because of Wilson’s disease. The process showed severe jaundice with fatigue, poor appetite and continually rising serum aminopherase. The blood serum was found positive for the anti-HEV IgG antibody but negative for a
View details →DOI: 10.3855/jidc.3638 Negative / Null Result ReportOpen accessEducation
Reza sharifinia, Abdolhussein Shakurnia · 2022 · Journal of Medical Education Development
Background & Objective: Critical thinking (CT) is essential in medical education and improves physicians' clinical ability and competency. The present study aimed to determine the CT level and the trend of its changes in medical education at Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran, in 2021. Materials & Methods: In this cross-sectional descriptive study, the CT levels of 165 medical students in three curricular phases, namely basic sciences, externship, and internship courses were examined. The data collection tool was the California Critical Thinking Skills Questionnaire
View details →DOI: 10.52547/edcj.15.45.26 Negative / Null Result ReportOpen accessComputer Science
Fırat Öncel, Matthias Bethge, Beyza Ermis et al. · 2024 · arXiv
In the last decade, the generalization and adaptation abilities of deep learning models were typically evaluated on fixed training and test distributions. Contrary to traditional deep learning, large language models (LLMs) are (i) even more overparameterized, (ii) trained on unlabeled text corpora curated from the Internet with minimal human intervention, and (iii) trained in an online fashion. These stark contrasts prevent researchers from transferring lessons learned on model generalization and adaptation in deep learning contexts to LLMs. To this end, our short paper introduces empirical ob
Negative / Null Result ReportOpen accessComputer Science
Piyawat Lertvittayakumjorn, David Kinney, Vinodkumar Prabhakaran et al. · 2025 · arXiv
Generative large language models (LLMs) have demonstrated gaps in diverse cultural awareness across the globe. We investigate the effect of retrieval augmented generation and search-grounding techniques on LLMs' ability to display familiarity with various national cultures. Specifically, we compare the performance of standard LLMs, LLMs augmented with retrievals from a bespoke knowledge base (i.e., KB grounding), and LLMs augmented with retrievals from a web search (i.e., search grounding) on multiple cultural awareness benchmarks. We find that search grounding significantly improves the LLM p
Negative / Null Result ReportOpen accessEngineering
Haoyang Li, Yuchen Hu, Chen Chen et al. · 2024 · arXiv
Deep neural network (DNN)-based speech enhancement (SE) usually uses conventional activation functions, which lack the expressiveness to capture complex multiscale structures needed for high-fidelity SE. Group-Rational KAN (GR-KAN), a variant of Kolmogorov-Arnold Networks (KAN), retains KAN's expressiveness while improving scalability on complex tasks. We adapt GR-KAN to existing DNN-based SE by replacing dense layers with GR-KAN layers in the time-frequency (T-F) domain MP-SENet and adapting GR-KAN's activations into the 1D CNN layers in the time-domain Demucs. Results on Voicebank-DEMAND sho
Negative / Null Result ReportOpen accessPhysiology
Chunwei Ma, Chunwei Ma, Yongcai Zhao et al. · 2022 · Frontiers in Physiology
Hypoxic training improves the microcirculation function of human skeletal muscle, but its mechanism is still unclear. Silent information regulator 2 homolog 3 (Sirt3) can improve mitochondrial function and oxidative status. We aimed to examine the role of Sirt3 in the process of hypoxic training, which affects skeletal muscle microcirculation. C57BL/6 mice were assigned to control (C), hypoxic training (HT), Sirt3 inhibitor 3-(1H-1,2,3-triazol-4-yl) pyridine (3-TYP), and 3-TYP + hypoxic training (3-TYP + HT) groups (n = 6/group). Sirt3 inhibition was induced by intraperitoneal injection of Sir
View details →DOI: 10.3389/fphys.2022.921763 Negative / Null Result ReportOpen accessComputer Science
Àlex Miranda-Pascual, Javier Parra-Arnau, Thorsten Strufe · 2026 · arXiv
Sampling is renowned for its privacy amplification in differential privacy (DP), and is often assumed to improve the utility of a DP mechanism by allowing a noise reduction. In this paper, we further show that this last assumption is flawed: When measuring utility at equal privacy levels, sampling as preprocessing consistently yields penalties due to utility loss from omitting records over all canonical DP mechanisms -- Laplace, Gaussian, exponential, and report noisy max -- , as well as recent applications of sampling, such as clustering. Extending this analysis, we investigate suppression as
Negative / Null Result ReportOpen accessPsychiatry
Michael Pascal Hengartner, Silvia ePassalacqua, Gisela eHeim et al. · 2016 · Frontiers in Psychiatry
Purpose: To evaluate the efficacy of a post-discharge intervention for psychiatric inpatients aimed at preventing hospital readmissions and at improving patients’ mental health and psychosocial functioning. Methods: RCT using parallel group block randomisation including 151 patients with ≤3 hospitalisations within the last three years, a GAF score ≤60, and aged 18–64 years, assessed at two psychiatric hospitals from the canton of Zurich, Switzerland, between September 2011 and February 2014. Primary outcomes were rate and duration of rehospitalisation; secondary outcomes were mental health and
View details →DOI: 10.3389/fpsyt.2016.00027 Negative / Null Result ReportOpen accessComputer Science
Hanbing Liu, Haoyang Li, Xiaokang Zhang et al. · 2025 · arXiv
Direct Preference Optimization (DPO) has proven effective in complex reasoning tasks like math word problems and code generation. However, when applied to Text-to-SQL datasets, it often fails to improve performance and can even degrade it. Our investigation reveals the root cause: unlike math and code tasks, which naturally integrate Chain-of-Thought (CoT) reasoning with DPO, Text-to-SQL datasets typically include only final answers (gold SQL queries) without detailed CoT solutions. By augmenting Text-to-SQL datasets with synthetic CoT solutions, we achieve, for the first time, consistent and
Negative / Null Result ReportOpen accessImmunologic diseases. Allergy
Shunchuan Zhang, Lauren E Springer, Han-Zhi Rao et al. · 2021 · PLoS Pathogens
Cytomegalovirus (CMV) causes clinically important diseases in immune compromised and immune immature individuals. Based largely on work in the mouse model of murine (M)CMV, there is a consensus that myeloid cells are important for disseminating CMV from the site of infection. In theory, such dissemination should expose CMV to cell-mediated immunity and thus necessitate evasion of T cells and NK cells. However, this hypothesis remains untested. We constructed a recombinant MCMV encoding target sites for the hematopoietic specific miRNA miR-142-3p in the essential viral gene IE3. This virus diss
View details →DOI: 10.1371/journal.ppat.1009255 Negative / Null Result ReportOpen accessMathematics
Oliver J. Hines, Christian L. Hines · 2026 · arXiv
Gradient boosted decision trees require a stopping rule to avoid overfitting. The standard rule monitors a validation loss and stops if the loss fails to improve for a fixed patience period. However, the patience parameter has no interpretable scale and validation losses can be noisy or implicitly defined by a user-specified gradient. We propose ScoreStop, a gradient-based early-stopping rule that casts the stopping decision at each iteration as a test of the null hypothesis that the current predictor is the population risk minimizer. We use a functional score test, computed on validation data
Negative / Null Result ReportOpen accessComputer Science
Zelu Deng, Yujie Zhong, Sheng Guo et al. · 2021 · arXiv
This work aims at improving instance retrieval with self-supervision. We find that fine-tuning using the recently developed self-supervised (SSL) learning methods, such as SimCLR and MoCo, fails to improve the performance of instance retrieval. In this work, we identify that the learnt representations for instance retrieval should be invariant to large variations in viewpoint and background etc., whereas self-augmented positives applied by the current SSL methods can not provide strong enough signals for learning robust instance-level representations. To overcome this problem, we propose InsCL
