Jiang Liu

Outperforming prior open models, Instella-MoE sets a new standard for efficiency and performance in language modeling with its innovative architecture and training methods.

PhysWave achieves spatially consistent audio generation by combining physics-based priors with a unified control framework, revolutionizing text-to-FOA applications.

By embedding near-periodic motion patterns directly into the deformation space, this method achieves state-of-the-art performance in 4D medical image interpolation.

Low-rank velocity fields can transform unsupervised 4D medical image interpolation, yielding anatomically coherent results that rival supervised methods.

Achieving over 42% recall in semantic video communication could redefine how we transmit meaning in bandwidth-limited networks.

Re-ranking can make or break user engagement, and GR2 boosts performance by over 18% by harnessing the power of LLMs in ways previously unexplored.

Ophthalmic VQA models can be made more accurate and transparent by explicitly grounding them in spatially-localized lesion evidence, a crucial step towards clinical interpretability.

Ruyi2.5 achieves comparable performance to Qwen3-VL on general multimodal benchmarks while significantly outperforming it in privacy-constrained surveillance, demonstrating the effectiveness of its edge-cloud architecture.

You can now get illustrative visual references from surveillance systems in sensitive environments without exposing raw images, thanks to a novel privacy-preserving perception framework.

Stop trading off fidelity for visual quality in super-resolution: a new network learns to directly optimize for human-preferred aesthetics.

A new unified LVLM, OmniCT, bridges the gap between slice-level detail and volumetric understanding in CT scans, outperforming existing methods by a significant margin.