Jun Zhang

EditVerse3D achieves high-fidelity 3D object edits using only coarse region specifications, outperforming traditional methods that require precise inputs.

FlexiSLM can operate at frame rates as low as 4.0 Hz while maintaining high-quality speech, effectively halving inference time compared to traditional models.

ReasoningLens turns the opaque reasoning of large models into clear, actionable insights, enabling researchers to pinpoint errors and optimize performance like never before.

Forecasting future coding tasks can yield a dataset that is 58.1% relevant to real-world software engineering needs, sidestepping the pitfalls of historical data replay.

Cold items can be effectively denoised using content similarity, leading to substantial performance boosts in recommendation systems.

Turns out, LLMs rely far more on raw code access than documentation when answering repository-level questions, challenging the assumption that documentation is the primary driver of code understanding.

MiniMax-M2 proves that massive parameter counts don't always translate to better agentic performance; strategic activation of a smaller subset can unlock frontier-level intelligence.

Forget slow, complex training: you can now distill diffusion models to just 4 steps and still beat the state-of-the-art in preference alignment, aesthetics, and composition.

Revitalizing target-specific control within latent-query architectures for sequential CTR prediction yields consistent performance gains across diverse datasets and backbones, especially when combined with a simple position-aware reference.

Seedance 2.0 leapfrogs existing models by unifying multi-modal inputs (text, image, audio, video) into a single architecture for generating high-quality, longer-duration audio-video content.

Style transfer just got a whole lot easier: MegaStyle's 1.4M image dataset, generated via text-to-image models, unlocks reliable style similarity measurement and generalizable style transfer.