| unchecked | Further analyses reveal that subnetworks found from different pre-training tend to yield diverse mask structures and perturbation sensitivities.arXiv 2012.06908 · ext:c291075732ddf581 | 0.55 | 4.95 | 0 | 0 | Computer Science | 6 Oct 2026 |
| unchecked | Comparing with other structure selection methods that may need thousands of trials or iterative fine-tuning, our method is trained fully end-to-end in one training pass without bells and whistles.arXiv 1707.01213 · ext:bb6ae4c59b519d96 | 0.55 | 4.91 | 0 | 0 | Computer Science | 7 Oct 2026 |
| unchecked | By forcing some of the factors to zero, we can safely remove the corresponding structures, thus prune the unimportant parts of a CNN.arXiv 1707.01213 · ext:d367a0c06a64413c | 0.55 | 4.91 | 0 | 0 | Computer Science | 7 Oct 2026 |
| supported | Notably, AlphaEvolve developed a search algorithm that found a procedure to multiply two $4 \times 4$ complex-valued matrices using $48$ scalar multiplications; offering the first improvement, after 56 years, over Strassen's algorithm in t…arXiv 2506.13131 · ext:259477645a3f495d | 0.71 | 4.91 | 0 | 1 | Computer Science | 6 Oct 2026 |
| unchecked | Overall, it indicates that theoretical models and empirical studies produce indecisive outcomes based on a variety of criteria such as the level of development of the sampled nations, the methodology utilized, data coverage, and the resear…DOI 10.26668/businessreview/2023.v8i4.1772 · conceptual · ext:36993ff5e2aae035 | 0.55 | 4.86 | 0 | 0 | Economics, Econometrics and Finance | 10 Oct 2026 |
| unchecked | Also, we determine that the biggest obstacle of using LLMs in Healthcare are fairness, accountability, transparency and ethics.arXiv 2310.05694 · conceptual · ext:34451fd3f3f84729 | 0.55 | 4.86 | 0 | 0 | Medicine | 8 Oct 2026 |
| unchecked | This shift encompasses a move from discriminative AI approaches to generative AI approaches, as well as a shift from model-centered methodologies to data-centered methodologies.arXiv 2310.05694 · conceptual · ext:103b9c48cd9b1623 | 0.55 | 4.86 | 0 | 0 | Medicine | 8 Oct 2026 |
| unchecked | The results reveal that two leading AI weather models, i.e. GraphCast and Pangu, are tied on the probabilistic CRPS metric even though the former outperforms the latter in deterministic scoring.arXiv 2401.15305 · ext:b13f228a8d4f2654 | 0.55 | 4.86 | 0 | 0 | Earth and Planetary Sciences | 7 Oct 2026 |
| unchecked | Surprisingly, we find models are capable of zero-shot generalization to tasks in languages they have never intentionally seen.arXiv 2211.01786 · ext:d5a8257c314f81ca | 0.55 | 4.86 | 0 | 0 | Computer Science | 7 Oct 2026 |
| unchecked | Finetuning on multilingual tasks with English prompts further improves performance on English and non-English tasks leading to various state-of-the-art zero-shot results.arXiv 2211.01786 · ext:5616a42c5ef2242d | 0.55 | 4.86 | 0 | 0 | Computer Science | 7 Oct 2026 |
| unchecked | We find finetuning large multilingual language models on English tasks with English prompts allows for task generalization to non-English languages that appear only in the pretraining corpus.arXiv 2211.01786 · ext:c934808f71474ae1 | 0.55 | 4.86 | 0 | 0 | Computer Science | 7 Oct 2026 |
| unchecked | Experiments on benchmark CIFAR-10, CIFAR-100, and ImageNet datasets have shown that our network architecture has superior generalization ability compared to the original residual networks.arXiv 1610.02915 · ext:24f381ea8ec94140 | 0.55 | 4.81 | 0 | 0 | Computer Science | 10 Oct 2026 |
| unchecked | We find that a targeted parameterization derived from a Padé expansion of $w$ is the most reliable of these, producing accurate reconstructions of $w(z)$, the expansion history $H(z)$, and cosmological parameters such as $H_0$ and $Ω_m$ fo…arXiv 2504.02028 · ext:6099762c51ee848d | 0.55 | 4.81 | 0 | 0 | Physics and Astronomy | 9 Oct 2026 |
| unchecked | Meanwhile, a distant, low-eccentricity 5 Earth-mass planet replicates the observed fraction of highly inclined scatterers, but the overall inclination distribution is more excited than observed.arXiv 1905.09286 · ext:9e71857db9157d30 | 0.55 | 4.81 | 0 | 0 | Physics and Astronomy | 9 Oct 2026 |
| unchecked | Although the number of highly inclined scatterers from the Oort Cloud is ~3 times less than observed, Oort cloud enrichment from the Sun's galactic migration or birth cluster could resolve this.arXiv 1905.09286 · ext:3f9a7e2bc4e7510a | 0.55 | 4.81 | 1 | 0 | Physics and Astronomy | 9 Oct 2026 |
| supported | Our method, which is based on clausal proof minimization, allowed us to compute several 553-vertex unit-distance graphs with chromatic number 5, while the smallest published unit-distance graph with chromatic number 5 has 1581 vertices.arXiv 1805.12181 · ext:33129411bee32407 | 0.71 | 4.81 | 0 | 1 | Computer Science | 8 Oct 2026 |
| supported | We find the model complies with harmful queries from free users 14% of the time, versus almost never for paid users.arXiv 2412.14093 · ext:13365b9934329605 | 0.71 | 4.81 | 0 | 0 | Social Sciences | 6 Oct 2026 |
| unchecked | We present a new fully general non-commutative solution with 23 multiplications and show that this solution is new and is NOT an equivalent variant of the Laderman's original solution.arXiv 1108.2830 · ext:719f16d62bf9966a | 0.55 | 4.81 | 0 | 1 | Computer Science | 6 Oct 2026 |
| unchecked | We further find that the examined AI models tend to underestimate both the frequency and intensity of record-breaking events, and they underpredict hot records and overestimate cold records with growing errors for larger record exceedance.DOI 10.1126/sciadv.aec1433 · ext:aba01a8625216166 | 0.55 | 4.77 | 0 | 0 | Earth and Planetary Sciences | 10 Oct 2026 |
| unchecked | We demonstrate that forecast errors in AI models are consistently larger for record-breaking heat, cold, and wind than in HRES across nearly all lead times.DOI 10.1126/sciadv.aec1433 · ext:ad18d4b0a2b4ce9a | 0.55 | 4.77 | 0 | 0 | Earth and Planetary Sciences | 10 Oct 2026 |
| unchecked | Here, we show that for record-breaking weather extremes, the physics-based numerical model High RESolution forecast (HRES) from the European Centre for Medium-Range Weather Forecasts still consistently outperforms state-of-the-art AI model…DOI 10.1126/sciadv.aec1433 · ext:e24100ab1e6ca053 | 0.55 | 4.77 | 0 | 0 | Earth and Planetary Sciences | 10 Oct 2026 |
| unchecked | Using a variational autoencoder framework, FuXi-ENS optimizes a loss function that combines the continuous ranked probability score (CRPS) with the Kullback-Leibler divergence, enabling flow-dependent perturbations.DOI 10.1126/sciadv.adu2854 · conceptual · ext:7c5b4e752dc4d23f | 0.55 | 4.77 | 0 | 0 | Earth and Planetary Sciences | 11 Oct 2026 |
| unchecked | Comprehensive evaluations demonstrate that FuXi-ENS outperforms the ECMWF ensemble in key forecast metrics such as CRPS and Brier score.DOI 10.1126/sciadv.adu2854 · ext:8ce71887198efb8a | 0.55 | 4.77 | 0 | 0 | Earth and Planetary Sciences | 11 Oct 2026 |
| unchecked | Unexpectedly, these toroidal cells resemble band-like cells rather than high grid score units.DOI 10.1016/j.isci.2023.108102 · ext:fd2dfb3e0768bcd1 | 0.55 | 4.75 | 0 | 0 | Neuroscience | 11 Oct 2026 |
| unchecked | Dimensionality reduction techniques reveal that a subset of cells manifest a persistent torus across environments.DOI 10.1016/j.isci.2023.108102 · ext:624202b410c66d3d | 0.55 | 4.75 | 0 | 0 | Neuroscience | 11 Oct 2026 |
| unchecked | Subsequent pruning studies reveal that toroidal cells are crucial for path integration while grid cells are not.DOI 10.1016/j.isci.2023.108102 · ext:adaade5f2aff6dde | 0.55 | 4.75 | 0 | 0 | Neuroscience | 11 Oct 2026 |
| unchecked | This provides ProGen with an unprecedented range of evolutionary sequence diversity and allows it to generate with fine-grained control as demonstrated by metrics based on primary sequence similarity, secondary structure accuracy, and conf…arXiv 2004.03497 · ext:b3db439140d2ceb4 | 0.55 | 4.75 | 0 | 0 | Biochemistry, Genetics and Molecular Biology | 9 Oct 2026 |
| unchecked | Our results highlight the fact that both classical U-shaped generalization curve and the recently observed double descent curve are not intrinsic properties of the model family.arXiv 2008.01036 · conceptual · ext:f85385696a20b109 | 0.55 | 4.75 | 0 | 0 | Computer Science | 9 Oct 2026 |
| unchecked | We show that the generalization curve can have an arbitrary number of peaks, and moreover, locations of those peaks can be explicitly controlled.arXiv 2008.01036 · conceptual · ext:c92b0e2173cab4e3 | 0.55 | 4.75 | 0 | 0 | Computer Science | 9 Oct 2026 |
| unchecked | A much wider range of surface ages are found for the large moons Callisto, Ganymede, Titan, and Europa (4.1, 3.4, 1.8, and 0.18 Gyr old, respectively).DOI 10.3847/psj/ad29f4 · ext:a5ef2178a3fe1f1c | 0.55 | 4.75 | 0 | 0 | Physics and Astronomy | 9 Oct 2026 |
| unchecked | Iapetus, Hyperion, Phoebe, and Oberon have surface ages that are a few Myr to a few tens of Myr younger than when Neptune entered the PKB (i.e., they are 4.52–4.53 Gyr old).DOI 10.3847/psj/ad29f4 · ext:150cf5ab71a76ea8 | 0.55 | 4.75 | 0 | 0 | Physics and Astronomy | 9 Oct 2026 |
| unchecked | A global fit to BAO, CMB, and calibrated SN data favors modified recombination over DDE.arXiv 2504.15274 · ext:f55ec6493942d97f | 0.55 | 4.75 | 0 | 0 | Physics and Astronomy | 9 Oct 2026 |
| unchecked | On ImageNet, our pruned ResNet-50 with 30% FLOPs reduced outperforms the original model.DOI 10.1109/tcsvt.2023.3248659 · ext:f8fc8f6b1ac02aea | 0.55 | 4.75 | 0 | 0 | Computer Science | 8 Oct 2026 |
| unchecked | Precisely, we argue that channels revealing similar feature information have functional overlap and that each such similarity group can be reduced to a few representatives with little impact on the representational power of the model.DOI 10.1109/tcsvt.2023.3248659 · ext:115e402989e92510 | 0.55 | 4.75 | 0 | 0 | Computer Science | 8 Oct 2026 |
| unchecked | In this report, we argue that there is a realistic possibility that some AI systems will be conscious and/or robustly agentic in the near future.arXiv 2411.00986 · conceptual · ext:b3686cf5ef625439 | 0.55 | 4.75 | 2 | 0 | Social Sciences | 7 Oct 2026 |
| unchecked | В случае, когда $r-s$ есть степень простого числа и $r\leq 2s+1$, удалось установить порядок $\alpha(\mathscr{G}(G(n,r,s), p))$ и $\chi(\mathscr{G}(G(n,r,s), p))$.DOI 10.4213/sm8344 · conceptual · ext:7a26562e309ccfb1 | 0.55 | 4.70 | 0 | 0 | Mathematics | 9 Oct 2026 |
| unchecked | Найдены нетривиальные нижние и верхние оценки числа независимости и хроматического числа таких графов.DOI 10.4213/sm8344 · conceptual · ext:54ee3915d69d1749 | 0.55 | 4.70 | 0 | 0 | Mathematics | 9 Oct 2026 |
| unchecked | The time-to-solution for training FourCastNet measured on JUWELS Booster on 3,072GPUs is 67.4minutes, resulting in an 80,000times faster time-to-solution relative to state-of-the-art NWP, in inference.arXiv 2208.05419 · ext:3a64f649ca092564 | 0.55 | 4.70 | 0 | 0 | Earth and Planetary Sciences | 9 Oct 2026 |
| unchecked | FourCast-Net is optimized and scales efficiently on three supercomputing systems: Selene, Perlmutter, and JUWELS Booster up to 3,808 NVIDIA A100 GPUs, attaining 140.8 petaFLOPS in mixed precision (11.9%of peak at that scale).arXiv 2208.05419 · ext:1b06d3af0d636f57 | 0.55 | 4.70 | 0 | 0 | Earth and Planetary Sciences | 9 Oct 2026 |
| unchecked | We report that a data-driven deep learning Earth system emulator, FourCastNet, can predict global weather and generate medium-range forecasts five orders-of-magnitude faster than NWP while approaching state-of-the-art accuracy.arXiv 2208.05419 · ext:66e419929b00d1b1 | 0.55 | 4.70 | 0 | 0 | Earth and Planetary Sciences | 9 Oct 2026 |
| unchecked | Using a hybrid approach building on work by Laumann et al. we greatly extend the known satisfiable region for random k-QSAT to a density of $Ω(2^k/k^2)$.arXiv 0911.1696 · conceptual · ext:f1906785b0b94772 | 0.55 | 4.70 | 0 | 0 | Computer Science | 9 Oct 2026 |
| unchecked | We show that the LLL extends to a much more general geometric setting, where events are replaced with subspaces and probability is replaced with relative dimension, which allows to lower bound the dimension of the intersection of vector sp…arXiv 0911.1696 · conceptual · ext:f00b3832f3dc13a0 | 0.55 | 4.70 | 0 | 0 | Computer Science | 9 Oct 2026 |
| unchecked | The performance of the AOA algorithm is evaluated against some well-known metaheuristic algorithms, including Genetic Algorithm (GA), Differential Evolution (DE), Tabu Search (TS), Firefly Algorithm (FA), Bat Algorithm (BA), Whale Optimiza…DOI 10.1007/s11831-025-10451-0 · ext:2fdd2a55972ad637 | 0.55 | 4.65 | 0 | 0 | Computer Science | 11 Oct 2026 |
| supported | The border rank of the matrix multiplication operator for n by n matrices is a standard measure of its complexity. Using techniques from algebraic geometry and representation theory, we show the border rank is at least 2n^2-n.arXiv 1112.6007 · ext:72c84a0906e77f5f | 0.71 | 4.64 | 0 | 0 | Mathematics | 10 Oct 2026 |
| unchecked | Crafting the ideal Earth foundation model, we define eleven features which would allow such a foundation model to be beneficial for any geoscientific downstream application in an environmental- and human-centric manner.DOI 10.1038/s43247-025-03127-x · conceptual · ext:b3b2b4e8a810ff4a | 0.55 | 4.64 | 0 | 0 | Computer Science | 10 Oct 2026 |
| unchecked | We concretely apply this criterion to local Hamiltonians on various regular lattices, while bringing to bear the tools of spin glass physics which permit us to obtain new bounds on the SAT/UNSAT transition in random quantum satisfiability.arXiv 1509.07766 · conceptual · ext:e52b87b4efe473ef | 0.55 | 4.64 | 0 | 0 | Computer Science | 9 Oct 2026 |
| unchecked | Remarkably, evaluating this condition proceeds via a fully classical analysis of a hard-core lattice gas at negative fugacity on the Hamiltonian's interaction graph which, as a statistical mechanics problem, is of interest in its own right.arXiv 1509.07766 · conceptual · ext:626808e8ae4e775c | 0.55 | 4.64 | 0 | 0 | Computer Science | 9 Oct 2026 |
| unchecked | For the uniform probability model, we prove that the phase transition is easy in the sense that there is a polynomial algorithm that can solve a random instance of the problem with the probability asymptotic to 1 as the problem size tends…arXiv 1106.4575 · ext:c575f27d94c58d71 | 0.55 | 4.64 | 0 | 0 | Computer Science | 8 Oct 2026 |
| unchecked | For the fixed ratio model, we establish several upper bounds for the solubility threshold, and prove that random instances with parameters above these upper bounds can be solved polynomially.arXiv 1106.4575 · conceptual · ext:01e1853998d2c7ed | 0.55 | 4.64 | 0 | 0 | Computer Science | 8 Oct 2026 |
| unchecked | Finally, DESY5+DESI DR2+CC datasets provide nearly decisive evidence favoring the PAge model over the standard $Λ\mathrm{CDM}$ model.arXiv 2505.22369 · ext:dd83cac4ee6ce7df | 0.55 | 4.64 | 0 | 0 | Physics and Astronomy | 7 Oct 2026 |