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.
1354 papers reviewed by authors · 861–870 / 1354
This paper demonstrates that training lightweight, frozen adapters on interpretability artifacts enables large language models to generate reliable, high-quality self-interpretations across tasks and scales without modifying the underlying model.
This paper provides a rigorous probabilistic analysis of deep Leaky ReLU networks to derive a Lyapunov exponent governing activation stability, revealing limitations in standard initialization methods and proposing a novel "Lyapunov initialization" that sets this exponent to zero to ensure optimal training stability.
This study employs DFT+U calculations to systematically investigate the electronic, magnetic, and optoelectronic properties of the RE2NiMnO6 double-perovskite series, revealing how lanthanide contraction-induced octahedral distortions and the specific treatment of RE 4f electrons collectively govern the material's spin-channel asymmetry and functional potential.
This paper proposes a fixed-order perturbative on-shell construction of scattering amplitudes using BCFW-generated trees and integrated vacuum pairs, successfully reproducing the known one- and two-loop all-plus four-gluon amplitudes up to order through a polygon-organized inclusion-exclusion framework.
This paper presents a comprehensive framework demonstrating that primordial black holes captured by stars can either quietly consume the host or trigger explosive stellar disruption via accretion disk formation, producing distinct transient signals that offer a new probe for asteroid-mass dark matter and subsolar black hole mergers.
This paper introduces EURO-5K, a specialized dataset for extracting EU reporting obligations, and demonstrates that while legal pretraining offers marginal gains for fully fine-tuned models, it significantly boosts the performance of parameter-efficient tuning and accelerates learning with limited data, ultimately validating both discriminative and generative approaches for regulatory compliance automation.
This paper introduces two data-augmentation strategies—fractional pseudo-games and phantom players—to provide intuitive, interpretable regularization for paired comparison models, effectively stabilizing estimates and resolving non-identifiability issues while closely replicating the results of standard ridge regularization.
This paper proposes a compositional governance framework that introduces formal primitives for recursive delegation, contextual boundaries, and dynamic scoping to overcome the limitations of traditional authorization systems in managing the complex, autonomous nature of agentic AI.
This paper investigates the diagonal condition in the ring of Gaussian integers and characterizes the specific Gaussian integers for which the quotient ring satisfies this condition, meaning its multiplication table contains the identity element 1 exclusively on the main diagonal.
This paper introduces BaltiVoice, the first publicly available speech corpus and fine-tuned Whisper ASR model for the Balti language, which significantly reduces word error rates from a 182.18% zero-shot baseline to 30.07% on a 16.8-hour dataset derived from Mozilla Common Voice.