…By preventing conflicting signals from canceling each other out, this masking strategy preserves and enhances the magnitude of effective RL advantages, thereby significantly accelerating learning efficiency. Furthermore, we introduce query-level reweighting…
…Injecting 3D Spatial Priors into Vision-Language Models for Enhanced Geometric Reasoning (2026) Enhancing MLLM Spatial Understanding via Active 3D Scene Exploration for Multi-Perspective Reasoning (2026) Beyond Localization: A Comprehensive Diagnosis…
…https://omerbenishu.github.io/PhyGenHOI/ View arXiv page View PDF Project page Add to collection Community PhyGenHOI tackles the lack of physical realism in 4D generation. We model humans via MDM and…
…Technology changes humanity This breakdown captures how rapidly ASR is evolving, especially the clear divide between LLM-enhanced accuracy and the blazing throughput of CTC/TDT systems. The post also highlights the…
…collection Community Introduces the first framework-agnostic method to leverage a learned quality prediction model to guide semi-supervised learning (SSL). Two complementary mechanisms for integrating as drop-in enhancements to existing…
…graphs with enhanced fidelity to physical invariants . Our code is available at https://github.com/ContinuumCoder/Hodge-Spectral-Duality View arXiv page View PDF GitHub 2 Add to collection Community In this…
…Caijun Xu , , , Abstract DenoiseRL is a reinforcement learning framework that enhances reasoning in large language models by learning from incorrect traces through failure-oriented optimization, improving scalability and reducing dependence on external…
…Extensive experiments demonstrate that UniDDT achieves effective unification of multimodal understanding and generation with enhanced semantic consistency and scalability. For visual generation tasks, our UniDDT achieves 0.87 GenEval score and 86…
…Experiments demonstrate that PBSD consistently enhances performance across both in-domain and out-of-domain settings, and effectively transfers knowledge from short-context training to long-context inference, suggesting that its fine…
…to collection Community On-Policy Distillation (OPD) is a fundamental technique for efficient post-training of large language models (LLMs), with broad applications in agent learning, multi-task enhancement, and model compression…