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 ReportOpen accessComputer Science
Mike Zhang, Ali Basirat, Desmond Elliott · 2026 · arXiv
Prior work establishes that controlled contrastiveness between self-generated responses from large language models, set via reward scores, improves downstream preference tuning in English. We extend this method to multiple languages and evaluate two models across a total of 14 high and low-resource languages on a diverse set of tasks. Our central finding is that cross-lingual contrastive preference tuning on self-generations (CroCo) transfers without language-specific preference annotation. A reward model trained on English preferences (atop a multilingual base) produces useful within-language
Negative / Null Result ReportOpen accessComputer Science
Zhiwei Jia, Xuanlin Li, Zhan Ling et al. · 2022 · arXiv
Generalization in deep reinforcement learning over unseen environment variations usually requires policy learning over a large set of diverse training variations. We empirically observe that an agent trained on many variations (a generalist) tends to learn faster at the beginning, yet its performance plateaus at a less optimal level for a long time. In contrast, an agent trained only on a few variations (a specialist) can often achieve high returns under a limited computational budget. To have the best of both worlds, we propose a novel generalist-specialist training framework. Specifically, w
Negative / Null Result ReportOpen accessComputer Science
Lily H. Zhang, Rajesh Ranganath · 2023 · arXiv
Methods which utilize the outputs or feature representations of predictive models have emerged as promising approaches for out-of-distribution (OOD) detection of image inputs. However, these methods struggle to detect OOD inputs that share nuisance values (e.g. background) with in-distribution inputs. The detection of shared-nuisance out-of-distribution (SN-OOD) inputs is particularly relevant in real-world applications, as anomalies and in-distribution inputs tend to be captured in the same settings during deployment. In this work, we provide a possible explanation for SN-OOD detection failur
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 accessMathematics
Ziling Ma, Ángel López Oriona, Hernando Ombao et al. · 2026 · arXiv
We study adaptive pooling under predictive heterogeneity in high-dimensional multivariate time series forecasting, where global models improve statistical efficiency but may fail to capture heterogeneous predictive structure, while naive specialization can induce negative transfer. We formulate adaptive pooling as a statistical decision problem and propose a validation-driven framework that determines when and how specialization should be applied. Rather than grouping series based on representation similarity, we define partitions through out-of-sample predictive performance, thereby aligning
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 accessComputer Science
Minwu Kim, Anubhav Shrestha, Safal Shrestha et al. · 2025 · arXiv
Recent studies have shown that reinforcement learning with verifiable rewards (RLVR) enhances overall accuracy (pass@1) but often fails to improve capability (pass@k) of LLMs in reasoning tasks, while distillation can improve both. In this paper, we investigate the mechanisms behind these phenomena. First, we demonstrate that RLVR struggles to improve capability as it focuses on improving the accuracy of the easier questions to the detriment of the accuracy of the most difficult questions. Second, we show that RLVR does not merely increase the success probability for the easier questions, but
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 accessEngineering
Jiaqi Wu, Jingyi Yuan, Yang Weng et al. · 2025 · arXiv
Power system voltage regulation is crucial to maintain power quality while integrating intermittent renewable resources in distribution grids. However, the system model on the grid edge is often unknown, making it difficult to model physical equations for optimal control. Therefore, previous work proposes structured data-driven methods like input convex neural networks (ICNN) for "optimal" control without relying on a physical model. While ICNNs offer theoretical guarantees based on restrictive assumptions of non-negative neural network parameters, can one improve the approximation power with
Negative / Null Result ReportOpen accessEngineering
Kwon Byung-Ki, Oh Hyun-Bin, Kim Jun-Seong et al. · 2023 · arXiv
Video motion magnification amplifies invisible small motions to be perceptible, which provides humans with a spatially dense and holistic understanding of small motions in the scene of interest. This is based on the premise that magnifying small motions enhances the legibility of motions. In the real world, however, vibrating objects often possess convoluted systems that have complex natural frequencies, modes, and directions. Existing motion magnification often fails to improve legibility since the intricate motions still retain complex characteristics even after being magnified, which may di
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
Failed Experiment ReportOpen accessComputer Science
Zixian Huang, Kaichen Yang, Xu Huang et al. · 2026 · arXiv
A widely adopted strategy for model enhancement is to use synthetic data generated by a stronger model for supervised fine-tuning (SFT). However, for emerging reasoning models like Qwen3-8B, this approach often fails to improve reasoning capabilities and can even lead to a substantial drop in performance. In this work, we identify substantial stylistic divergence between teacher generated data and the distribution of student as a major factor impacting SFT. To bridge this gap, we propose a Teacher-Student Cooperation Data Synthesis framework (TESSY), which interleaves teacher and student model
Negative / Null Result ReportOpen accessComputer Science
Saikat Mondal, Chanchal K. Roy · 2025 · arXiv
High-quality answers in technical Q&A platforms like Stack Overflow (SO) are crucial as they directly influence software development practices. Poor-quality answers can introduce inefficiencies, bugs, and security vulnerabilities, and thus increase maintenance costs and technical debt in production software. To improve content quality, SO allows collaborative editing, where users revise answers to enhance clarity, correctness, and formatting. Several studies have examined rejected edits and identified the causes of rejection. However, prior research has not systematically assessed whether acce
Negative / Null Result ReportMedicine
Alshaibani, Kamadjaja, Sitalaksmi et al. · 2026 · Journal of molecular histology
Tooth extraction is a common procedure often followed by alveolar bone resorption, which may compromise future implant placement, prosthetic rehabilitation, esthetics, and periodontal support. Hydroxyapatite-chitosan (HA-Chi) scaffolds…
View details →DOI: 10.1007/s10735-026-10874-4 Negative / Null Result ReportOpen accessComputer Science
Hongyu Zhou, Yinan Zhang, Aixin Sun et al. · 2025 · arXiv
Multimodal recommendation systems are increasingly popular for their potential to improve performance by integrating diverse data types. However, the actual benefits of this integration remain unclear, raising questions about when and how it truly enhances recommendations. In this paper, we propose a structured evaluation framework to systematically assess multimodal recommendations across four dimensions: Comparative Efficiency, Recommendation Tasks, Recommendation Stages, and Multimodal Data Integration. We benchmark a set of reproducible multimodal models against strong traditional baseline
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 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 ReportMedicine
Baudin, Pouyau, Subtil et al. · 2026 · JAMA
Prone positioning has been shown to improve respiratory mechanics and oxygenation, but its clinical benefit in infants with acute viral bronchiolitis receiving high-flow nasal cannula (HFNC) support remains unknown. To investigate whether…
View details →DOI: 10.1001/jama.2026.11078 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 ReportMedicine
Parirokh, Manochehrifar, Nakahee et al. · 2026 · Iranian endodontic journal
The close relationship between pulp stones in the pulp chamber and pulp neurovascular tissues suggests that the presence of pulp stones may compromise successful anesthesia. The present study assessed the effect of pulp stone presence on…
View details →DOI: 10.22037/iej.v21i1.46586 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 ReportMedicine
Ziltzer, Bulbul, Barry et al. · 2026 · The Laryngoscope
To investigate whether routine involvement of the medical hospitalist service (MHS) in postoperative medical management of head and neck cancer (HNC) patients undergoing major ablative head and neck surgery with free tissue transfer (FTT)…
View details →DOI: 10.1002/lary.70659 Negative / Null Result ReportOpen accessPhysics
Martin Amft, Natalia V. Skorodumova · 2011 · arXiv
The present density functional theory study addresses the question whether the presence of H2O influences the catalytic activity of small gold clusters, Au1-4/MgO(100), towards the oxidation of carbon monoxide. To this end, we studied the (co-)adsorption of H2O and CO/O2 on these gold clusters. The ground state structures in the presence of all three molecular species, that we found, are Au1O2/MgO and Au2-4CO/MgO with H2O adsorbed on the surface in the proximity of the clusters-molecule complex. In this configuration the catalytic activity of Au1-4/MgO is indifferent to the presence of H2O. We
Negative / Null Result ReportMedicine
Bakhshalipoor, Bolandi, Nahidi et al. · 2026 · Scientific reports
This study evaluated the effects of double nanoemulsions of saffron petal extract (DN-SPE), stabilized with soybean (SPI) and pea protein isolates (PPI), alone or combined with Lepidium sativum L. seed gum (LSG) (SL or PL), on the quality…
View details →DOI: 10.1038/s41598-026-52693-3 Negative / Null Result ReportMedicine
Wang, Li, Zhang et al. · 2026 · Child development
This study examines how center-based parenting interventions aimed at improving early child development in rural China affect the mental health of caregivers. Data from an analytic sample of 615 caregiver-child dyads (children aged 6 to 24…
View details →DOI: 10.1093/chidev/aacag099 Negative / Null Result ReportMedicine
Eminoğlu · 2026 · Food science & nutrition
In this study, yogurt formulations with paraprobiotic (heat inactivated Lactobacillus acidophilus ) and postbiotic (autolyzed and dried Saccharomyces cerevisiae ) additions were developed to improve the techno-functional properties of…
View details →DOI: 10.1002/fsn3.71787 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
Negative / Null Result ReportOpen accessComputer Science
Arijit Sehanobish, Charles Lovering · 2026 · arXiv
We study the \textit{parameter placement problem}: given a fixed budget of $k$ trainable entries within the B matrix of a LoRA adapter (A frozen), does the choice of which $k$ matter? Under supervised fine-tuning, random and informed subsets achieve comparable performance. Under GRPO on base models, random placement fails to improve over the base model, while gradient-informed placement recovers standard LoRA accuracy. This regime dependence traces to gradient structure: SFT gradients are low-rank and directionally stable, so any subset accumulates coherent updates; GRPO gradients are high-ran
