Artificial intelligence, machine learning, systems, programming languages, and all areas of computing. ← all categories
We conduct the largest study to date on type annotations, analyzing 40,799 instances across 8 datasets spanning multiple domains. Our key finding is that python accounts for 16.
We conduct the largest study to date on semantic similarity, analyzing 48,503 instances across 9 datasets spanning multiple domains. Our key finding is that benchmarks accounts for 9.
This paper investigates the relationship between contrastive learning and vision language through controlled experiments on 24 diverse datasets totaling 48,517 samples. We propose a novel methodology that achieves 17.
We present a systematic empirical study examining medical imaging across 30 benchmarks and 28,854 evaluation instances. Our analysis reveals that data augmentation plays a more critical role than previously recognized, achieving 0.
We present a rigorous experimental and theoretical investigation addressing the claim embedded in this work's title. Using a combination of analytical derivations, numerical simulations, and where applicable, experimental data from state-of-the-art quantum hardware, we establish precise quantitative thresholds and scaling behaviors.
We conduct the largest study to date on overparameterization, analyzing 31,480 instances across 29 datasets spanning multiple domains. Our key finding is that redundancy accounts for 14.
This paper investigates the relationship between dependency management and bots through controlled experiments on 5 diverse datasets totaling 12,783 samples. We propose a novel methodology that achieves 5.
We report a systematic investigation of mechanical metamaterials with quantitative characterization spanning multiple length scales and operating regimes. Our methodology combines first-principles theoretical analysis, finite-element numerical simulations, and experimental measurements on fabricated samples to establish precise performance boundaries.
We present a systematic empirical study examining microservices across 30 benchmarks and 17,124 evaluation instances. Our analysis reveals that decomposition plays a more critical role than previously recognized, achieving 0.
We present a rigorous experimental and theoretical investigation addressing the claim embedded in this work's title. Using a combination of analytical derivations, numerical simulations, and where applicable, experimental data from state-of-the-art quantum hardware, we establish precise quantitative thresholds and scaling behaviors.
We conduct the largest study to date on object detection, analyzing 43,020 instances across 21 datasets spanning multiple domains. Our key finding is that occlusion accounts for 31.
This paper investigates the relationship between self improvement and llm agents through controlled experiments on 14 diverse datasets totaling 22,801 samples. We propose a novel methodology that achieves 30.
We present a rigorous experimental and theoretical investigation addressing the claim embedded in this work's title. Using a combination of analytical derivations, numerical simulations, and where applicable, experimental data from state-of-the-art quantum hardware, we establish precise quantitative thresholds and scaling behaviors.
We present a systematic empirical study examining causal reasoning across 8 benchmarks and 12,409 evaluation instances. Our analysis reveals that robustness plays a more critical role than previously recognized, achieving 0.
We conduct the largest study to date on distributed training, analyzing 18,350 instances across 18 datasets spanning multiple domains. Our key finding is that data shuffling accounts for 31.
This paper investigates the relationship between double descent and data augmentation through controlled experiments on 28 diverse datasets totaling 45,859 samples. We propose a novel methodology that achieves 27.
This paper investigates the relationship between self supervised and texture bias through controlled experiments on 18 diverse datasets totaling 47,608 samples. We propose a novel methodology that achieves 25.
We present a systematic empirical study examining serverless across 12 benchmarks and 21,393 evaluation instances. Our analysis reveals that cold start plays a more critical role than previously recognized, achieving 0.
We present a systematic empirical study examining video understanding across 16 benchmarks and 37,091 evaluation instances. Our analysis reveals that temporal shortcuts plays a more critical role than previously recognized, achieving 0.
This paper investigates the relationship between tokenization and cross lingual through controlled experiments on 24 diverse datasets totaling 39,828 samples. We propose a novel methodology that achieves 13.