A Confident Learning framework adapted to segmentation for auditing label bias directly in the training data without a clean, unbiased ground truth is presented, and it is shown that label bias influences subgroup separability in the encoder's feature space, an artifact the authors leverage for bias mitigation rather t...
Aditya Parikh, Stella Frank, Snehayan Das et al.· arXiv.org· 2 citations
Post-training quantization (PTQ) is a practical path to deploy large diffusion models, but quantization noise can accumulate over the denoising trajectory and degrade generation quality. We propose Q-Drift, a sampler-side correction that aims to preserve the intended sampling marginals through a deterministic drift adj...
Goal-Driven Data Optimization (GDO) is presented, a framework that computes six sample descriptors for each candidate and constructs optimized 1$\times$ training subsets for different goals that enables faster convergence with fewer training samples under a fixed training protocol.
Ru-Jie Wu, Haozhe Zhao, Hai Ci et al.· arXiv.org· 0 citations
Pose-VLA is proposed, a decoupled paradigm that separates VLA training into a pre-training phase for extracting universal 3D spatial priors in a unified camera-centric space, and a post-training phase for efficient embodiment alignment within robot-specific action space.
Coral reefs are on the brink of collapse, with climate change, ocean acidification, and pollution leading to a projected 70-90% loss of coral species within the next decade. Reef restoration is crucial, but its success hinges on introducing automation to upscale efforts. In this work, we present a highly configurable A...
Scarlett Raine, Emilio Olivastri, Benjamin Moshirian et al.· arXiv.org· 3 citations
Squeeze3D is a novel framework that leverages implicit prior knowledge learnt by existing pre-trained encoders and decoders to compress 3D data at extremely high compression ratios and can flexibly support different formats, including meshes, point clouds, and radiance fields.
Rishit Dagli, Yu-Shi Guan, Sankeerth Durvasula et al.· 0 citations
This work proposes a single-stage, EM-style framework for generative noisy-label learning that is direction-agnostic and avoids explicit image synthesis, and introduces Partial-Label Supervision (PLS), an instance specific prior over clean labels that balances coverage and uncertainty, improving data-dependent regulari...
Feng-Bei Liu, Chong Wang, Yuanhong Chen et al.· IEEE Transactions on Pattern...· 2 citations
Granularity alignment, not algorithm choice, localizes when routing helps, and what category supervision buys when deployed through routing is quantified.
Fast LeWorldModel (Fast-LeWM), a fast latent world model that replaces repeated local rollout with action-prefix prediction, improves average success over LeWM while substantially reducing planning time, achieving lower open-loop latent loss whose growth becomes significantly slower as the rollout horizon increases.
This work introduces MulTaBench, a benchmark of 40 datasets, split equally between image-tabular and text-tabular tasks, designed to enable the research of novel architectures which incorporate joint modeling and target-aware representations, paving the way for the development of novel Multimodal Tabular Foundation Mod...
Alan Arazi, Eilam Shapira, Shoham Grunblat et al.· arXiv.org· 3 citations
Targeted data selection seeks training examples from a candidate pool that improve downstream task performance. While trajectory-based selectors effectively guide this process, existing approaches typically construct reference states by warming up on the candidate pool itself, thereby inheriting pool-dependent distribu...
Huitao Yang, Hengzhi He, Tung Sum Thomas Kwok et al.· 0 citations
This work finds that hallucinations often arise when the model relies less on perceptual inputs, and that changing this reliance affects its predictions, and proposes a training-free framework that shifts relevance toward perceptual tokens by optimizing key-value representations during decoding, without modifying model...
Itai Allouche, Joseph Keshet· arXiv.org· 1 citation
Professor Paris Smaragdis develops computers with superhuman hearing for applications like speech recognition, music understanding, and environmental monitoring.
Exploring how generative AI could make machine vision more accessible to businesses. The post GenEye in a Box: Making Machine Vision Something You Can Just Ask For appeared first on GPT-Lab.
Jennifer Neville did not want to go into computer science—but that’s exactly where she landed. Neville discusses the starts and stops that led to her professional sweet spot and her work identifying “surprising failures” making it hard for AI to handle complexity. The post What AI gets wrong and what failure teaches us appeared first on Microsoft Research.
Computer scientist, entrepreneur, and philanthropist will collaborate with the MIT Schwarzman College of Computing to advance AI and scientific discovery.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
