Papers where the original authors have engaged with our plain-language explanation.
For every paper on this page, at least one of the original authors has seen our plain-language explanation and engaged with it — either confirming it reads accurately or requesting corrections that we then applied. An endorsement does not mean the authors formally approve every sentence, but it does mean the explanation has passed the eyes of the people who wrote the paper.
1359 papers reviewed by authors · 921–930 / 1359
BIRDNet is an interpretable deep neural network that mines Boolean implication relationships from tabular data to construct a sparse, structurally constrained architecture, achieving competitive performance with significantly fewer parameters while enabling direct extraction of biologically meaningful rules without external priors.
This paper empirically audits the k-NAF budget-accounting mechanism in Anchored Decoding across fixed and adaptive workloads, finding that cumulative KL spend consistently remains well below sequence-level budgets and that instances of apparent budget exhaustion are attributable to proxy artifacts rather than actual mechanism failures.
This paper introduces hybrid neural world models, a single-network framework that predicts physical dynamics with significant speedups over classical solvers while implicitly generating an error map to detect sharp discontinuities like shocks and contacts, enabling a fallback mechanism that substantially reduces prediction errors without requiring additional calibration or governing-equation knowledge.
Using GASTRO library simulations, this study demonstrates that early high-density star-forming clumps or retrograde mergers can suppress star formation to create the Milky Way's observed -rich and -poor chemical bimodality, whereas prograde mergers fail to do so, thereby supporting clumpy disk galaxies at as likely progenitors of our Galaxy.
This paper proposes a history-aware adaptive reduced-order modeling framework using incremental singular value decomposition (iSVD) that dynamically updates basis functions via occasional full-order corrections, demonstrating superior predictive accuracy and computational efficiency over existing methods for complex nonlinear problems like the Burgers equation, Sod shock tube, and rotating detonation engines.
This paper introduces the Frequency Guidance Operator (FGO), a novel algorithm that enhances diffusion-based visuomotor policies by steering action generation through intermediate sub-frequency manifolds to suppress high-frequency noise from human demonstrations while preserving essential task details.
This paper demonstrates that GPU scan performance bottlenecks in Parquet files stem from suboptimal, CPU-centric configurations rather than the format itself, and shows that applying GPU-aware settings can boost effective read bandwidth to 125 GB/s without altering the Parquet specification.
This paper reviews the physical mechanisms and bottlenecks of muon-catalyzed fusion, proposes a synergistic four-step scheme to overcome alpha-sticking and achieve energy gain, and outlines a conceptual fusion-fission hybrid reactor design for breeding plutonium.
This paper introduces a reverse-martingale regularized recurrent neural network (RMRNN) that integrates a backward-coherence penalty and a Shiryaev–Roberts detector to simultaneously improve probabilistic precipitation forecasts and provide significantly earlier, more reliable drought and flood warnings across diverse global regions compared to existing baselines.
Analysis of Kepler photometry reveals that stars hosting confirmed exoplanets exhibit systematically higher rotational modulation dispersion and a unique bimodal distribution in their magnetic activity coherence compared to non-planet-hosting stars, suggesting that planetary systems may influence the temporal organization of stellar magnetic activity and dynamos.