Author

Xing-Yu Shen

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Preprint Oct 2026

Latent Compaction with Test-Time Training

Results show that treating the KV cache as a neural network and optimizing it with test-time training can preserve retrieval accuracy under aggressive compression, and transfers across model scales and families.

Yingfa Chen, Xing-Yu Shen, Chao-Jun Xiao et al. · 1 citation · ⚡1