Papers, modelos e datasets em alta no Hugging Face, além do blog oficial — com leitura editorial em português.
Egocentric human video can effectively replace teleoperated robot trajectories for embodied model pretraining, achieving better performance with reduced data collection costs.
FreeStyle is a scalable dual-reference generation framework that uses community LoRA mining to create large-scale style-content triplets while addressing content leakage through di…
MobileForge enables efficient adaptation of mobile GUI agents through annotation-free learning by combining real app interaction grounding with hierarchical feedback-guided policy…
LEDGERAGENT is a method for customer service agents that maintains task states in a separate ledger to improve policy adherence and state management during tool calling.
World Action Models are predictive-action systems that generate future states for decision-making, with designs balancing representational richness against computational constraint…
Go-with-the-Track unifies motion control and reference image compositing in video generation by using point-track embeddings with spatial-aware encoding and video diffusion transfo…
A self-supervised transfer learning approach for parking spot occupancy recognition that achieves high accuracy with minimal labeled data through two-stage training and deployment…
FlowBender is a closed-loop framework that addresses constraint satisfaction in diffusion and flow models by training networks to correct alignment errors using inference-time feed…
Grouped Query Experts (GQE) improves Transformer efficiency by selectively activating query heads based on token content while maintaining key-value cache benefits of grouped-query…
Multimodal large language models exhibit social bias driven by specific visual attributes, with fashion style and socioeconomic cues having the greatest impact on model judgments.
Large language models can be trained through reinforcement learning to develop a meta-capability enabling continuous learning and adaptation across long sequences of tasks in dynam…
CogniRoute is a schema-guided Mixture-of-Experts framework for social video question answering that improves multimodal reasoning through cognitive schema factorization and route-a…