…Evaluations on the LOL, SID, SMID, and SDSD benchmarks demonstrate overall improvements over Retinexformer and recent state-of-the-art methods. Code and pretrained weights are available at https://github.com/YoussefAboelwafa…
…and partial rollouts, throughput optimization , and replay/recomposition for self-improvement. We ground the framework with case studies in math, code/SQL, multimodal reasoning, tool-using agents, and agentic skill benchmarks that…
…LEAD reaches 53.36 acc / 3714 tokens / +0.68 AES, the only method that improves accuracy over base while reducing length. 📄 Paper: https://arxiv.org/abs/2605.09806 💻 Code: https://github.com…
…Maintains high-quality generation (gFID 1.05) with only 45 denoising steps, thanks to improved local continuity. Available Models: Pre-trained tokenizers based on DINOv2, SigLIP2, DINOv3, and MAE backbones! Code & Models…
…Experiments on four benchmarks demonstrate that TacoMAS outperforms nearly 20 multi-agent baselines, achieving an average improvement of 13.3% over the strongest baseline. The codes are released at https://github.com…
…Venkata Pushpak Teja Menta Abstract A self-contained text-to-speech to speech-to-text flywheel approach significantly improves niche-domain Indic automatic speech recognition performance through synthetic data generation and low…
…These gains carry over to RL-based post-training, where optimizing PhyMotion leads to larger and more consistent improvements than optimizing existing rewards, improving motion realism across both autoregressive and bidirectional video…
…Abstract AgensFlow is an open-source framework that treats multi-agent coordination as an online policy-learning problem under partial observability, enabling learned routing to improve coordination-heavy workflows over static approaches…
…evaluation code, ground-truth generation resources, and fine-tuning artifacts. Our results show that even strong frontier VLMs struggle substantially on the harder settings, while fine-tuned open models improve but remain…
…Ziyu Guo , , , Abstract ATLAS presents a visual reasoning framework that combines agentic operations and latent representations using functional tokens, enabling efficient training and improved performance on complex benchmarks. AI-generated summary Visual…