A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Sertaç Karaman, Emilio Frazzoli
Introduces PRM* and RRT*, provably asymptotically optimal sampling-based motion planning algorithms, with analysis via random geometric graphs.
Vijay Badrinarayanan, Alex Kendall, Roberto Cipolla
SegNet introduces a deep encoder-decoder architecture for semantic segmentation that uses max-pooling indices for efficient non-linear upsampling, achieving competitive accuracy with low memory and inference time.
Liangchen Li, Jiajun He
Bidirectional consistency models extend consistency models to enable both forward and reverse sampling, improving image generation quality and flexibility.
Unknown
AudioLCM introduces a latent consistency model for efficient text-to-audio generation, enabling high-quality synthesis in few sampling steps.
Unknown
The paper compares input-level test-time augmentation against output-level sampling for LLMs at matched compute, finding input diversity yields better accuracy.
Matthias Zwicker, Hanspeter Pfister, Jeroen van Baar, et al.
EWA splatting introduces a resampling filter framework using elliptical Gaussian kernels for high-quality, alias-free rendering of volume and point-sampled surface data.
Unknown
Introduces RAS, a training-free sampling strategy for Diffusion Transformers that dynamically adjusts token updates based on region-adaptive criteria.
Unknown
SelfCheckGPT is a zero-resource, black-box hallucination detection method for generative LLMs that uses sampling-based consistency checking.
Bohan Li, Yutai Hou, Wanxiang Che
This survey categorizes data augmentation methods in NLP into paraphrasing, noising, and sampling based on the diversity of augmented data, and reviews their applications and challenges.
L. Yang, Zhilong Zhang, Yang Song, et al.
A comprehensive survey of diffusion models, categorizing research into efficient sampling, improved likelihood estimation, and handling special data structures, with a review of applications across domains.
Debottam Dutta, Jaehoon Hahm, Jianchong Chen, et al.
TILT is a training-free framework that improves compositional text-to-image generation by using a model-intrinsic reward to guide diffusion sampling toward joint concept presence.
Rahul Raguram, Ondřej Chum, Marc Pollefeys, et al.
USAC extends RANSAC with a universal framework integrating advanced sampling, verification, and model refinement for robust estimation.