Search and filter through extracted claims from AI researchers.
Showing 1361-1380 of 2685 claims of type "fact"
A unified framework can enable a single cross-skill policy that performs strongly on grasping, relocation, in-hand rotation, and translation while generalizing to unseen objects
Model scale shows weak correlation with robustness to multi-attribute biases in vision-language models, with Spearman correlation dropping from 0.68 on ImageNet to only 0.05 on multi-attribute bias benchmarks
Training data quality is more important than model scale for VLM robustness, with curated datasets yielding up to 25% improvements over uncurated alternatives at comparable scale
Decoupling 3D geometric reconstruction from semantic integration enables more efficient open vocabulary 3D scene understanding from monocular video
State-of-the-art deep learning weather emulators rival or surpass physics-based forecasts in deterministic temperature skill at 10-15 day lead times, but do so at the cost of reduced spectral fidelity through blurring
Most deep learning weather emulators under-represent peak intensities of extreme heat events, with IFS recall greater than any of the emulators tested
Qwen-UI-Agent demonstrates a real-world centric foundation GUI agent spanning mobile, computer-use, web, and DeepSearch environments with unified action space and online RL supporting 100+ turn trajectories
A single LLM can orchestrate the whole knowledge retrieval process by planning complementary subqueries, executing searches, reading papers, and locating relevant evidence
Contrastive Denoising Autoencoder can refine BERT embeddings to achieve perturbation-invariant representation by jointly optimizing contrastive and reconstruction objectives
Collecting diverse egocentric data across scenes, objects, motions, and embodiments remains costly for embodied AI
Depth sensors remain underexplored for affordance segmentation in wearable robots despite their potential to complement RGB data
Augmenting 400 real trajectories with 400 synthetically generated egocentric manipulation videos improves visual fidelity, geometric stability, and action alignment in long egocentric rollouts
Integrating Sentinel-1 SAR and Sentinel-2 multispectral satellite data with street-level imagery from Mapillary can overcome cloud-induced temporal gaps and provide more accurate parcel-level agricultural monitoring
Semi-supervised methods using ensemble consensus can boost predictive accuracy across diverse molecular datasets, task types, and graph neural network architectures
An automated pipeline can successfully filter and curate crowdsourced street-level imagery at scale, producing 46,050 analysis-ready annotated images from over 900,000 initial images
Topoformer achieves state-of-the-art performance on graph classification and molecular property prediction benchmarks while being more scalable than traditional persistent homology methods
Encoding topological structure into attention-friendly sequences via Topo-Scan enables Transformers to capture multi-scale graph patterns without costly persistent homology diagram computations
Camera motion in video generation is primarily established during high-noise stages of the diffusion process, where coarse spatiotemporal structures are formed
Video re-shooting can be achieved without explicit 3D priors or paired training data by using text-driven semantic viewpoint specification and self-supervised learning of camera dynamics
Lower and middle Transformer layers build semantic representations while upper layers increasingly specialize them for prediction, creating a natural division of labor
Pipeline data may be stale or degraded.
Last synthesis: 2026-09-20. 8,986 pending.