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.
956 papers reviewed by authors · 501–510 / 956
This paper introduces Fair Fine-tuning (FFt), a method that mitigates distribution inference attacks by fine-tuning models on complementary data under Equalized Odds constraints, theoretically proving that adversarial advantage is bounded by fairness disparity and empirically demonstrating significant reductions in attack success across diverse datasets.
JenBridge is a novel, modular framework that leverages a Transformer-based generative model and an LLM-driven director agent to produce high-fidelity, long-form video soundtracks with natural, narrative-coherent transitions across scene changes, validated by the newly proposed LVS Benchmark.
This paper identifies "Marginal Path Collapse" as a critical failure mode in inference-time steering of diffusion models caused by mismatched noise schedules or negative exponents, and proposes the Adaptive Path Correction with Exponents (ACE) framework to mathematically guarantee path existence and significantly improve performance in complex compositional tasks like drug design and image generation.
This paper demonstrates that the quantitative openness profile of a nonlinear system's vector field imposes specific necessary lower bounds on the growth rate of stabilizing feedbacks, revealing that Brockett's topological condition is fundamentally governed by quantitative gain requirements rather than being merely a binary obstruction.
This study of 3,538 patients reveals that elevated body mass index and African ancestry are independently associated with a higher likelihood of developing the more aggressive Luminal B breast cancer subtype compared to Luminal A, with obesity partially mediating the link between ancestry and tumor phenotype.
This study demonstrates that the long-standing discrepancy between the observed Fundamental Plane of Early-Type Galaxies and hydrodynamical simulations like IllustrisTNG100-1 can be largely resolved by adopting observationally motivated measurement techniques and accounting for mass-dependent Initial Mass Function variations, highlighting the critical role of observational realism and stellar population modeling in interpreting galaxy scaling relations.
This systematic study on CIFAR-10 demonstrates that student capacity is a critical moderator of knowledge distillation effectiveness, revealing that larger students (R34) benefit significantly more than smaller ones (R18), while also highlighting the necessity of fixing implementation bugs and input-resolution mismatches to achieve optimal distillation performance.
This paper proposes a lightweight CNN-based anomaly detection framework for Spallation Neutron Source high-voltage converter modulators that leverages architectural inductive bias by strategically ordering temporal and cross-channel operations, achieving state-of-the-art performance in identifying fault precursors across multiple subsystems.
This paper introduces FEM-Bench, a structured benchmark based on computational mechanics tasks designed to rigorously evaluate the ability of large language models to generate scientifically valid finite element method code, revealing that even state-of-the-art models struggle to consistently solve these nontrivial problems.
This paper derives a universal sharp uncertainty relation, , that identically constrains the variances of position and momentum in classical light beams, coherent quantum states, and individual photons.