This is your work, valued
Shenzhen
Sparse-vLLM. A sparse-first inference engine (sparsevllm). It also contains DeltaKV compressor training + evaluation tooling (deltakv).
60LowRankClone. [NeurIPS 2025 Spotlight] A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone.
48MEFT. Python
22HRN. Code of Multi-Defendant Legal Judgment Prediction via Hierarchical Reasoning
15Uni-X. [ICLR 2026] Uni-X: Mitigating Modality Conflict with a Two-End-Separated Architecture for Unified Multimodal Models
10qydemo. qy
1UGen-Pro. Python
1Codex-Skills. Python
1 views searches