Yu-Chen Xie

INT4 KV caching for MLA models is now possible, achieving 3.59x compression at 128K context and enabling 5.168x higher throughput in cache-pressure serving workloads.

Multi-teacher distillation fails not because of bad routing, but because high-variance instruction feedback quietly drowns out precision domains like mathematics during shared student updates.

The deep variational framework not only recovers established smectic-A defect structures but also predicts novel configurations like a chevron-like smectic-C state in confined geometries.

RoomWright transforms indoor scene synthesis by prioritizing functional usage, enabling interactive environments that are ready for real-world robotic applications.

AFD-Ledger reveals that optimizing deployment for AFD can drastically cut evaluation costs while exposing the nuanced performance dynamics between homogeneous and heterogeneous setups.

LSA slashes indexing overhead while maintaining full attention performance, enabling efficient long-context processing for models with up to one million tokens.

Achieve full-attention accuracy with 10x operator speedup and 4.7x throughput improvement in long-context LLM inference by overlapping KV cache transfers with computation.

LongCat-Next shatters the language-centric paradigm by unifying text, vision, and audio into a single autoregressive model with minimal modality-specific design, finally reconciling understanding and generation in discrete vision modeling.