Author

Harry Dong

2 papers 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.

Preprint Aug 2026

Full-bandwidth transformer

This work trains 1B-parameter full-bandwidth transformers on up to 400B tokens and finds that latent feedback improves validation loss, 5-shot language-model evaluation, math and coding generation, and instruction-tuned performance.

Xi Wang, Ziyang Cai, Zheng Zhan et al. · 5 citations

#artificial intelligence Preprint Sep 2026

Almost Free State Prediction Separation

This paper takes state--prediction separation to its limit with a free pause token: a prediction stream that writes no keys or values at all and so rides the sequence's existing positions, and reduces the raw flops required at inference time.

John Langford, Nathan Godey, Giovanni Monea et al. · 1 citation