Multi-Token Prediction: Exploring Performance on NLP Benchmarks

Table of Links

Abstract and 1. Introduction

2. Method

3. Experiments on real data

4. Ablations on synthetic data

5. Why does it work? Some speculation

6. Related work

7. Conclusion, Impact statement, Environmental impact, Acknowledgements and References

A. Additional results on self-speculative decoding

B. Alternative architectures

C. Training speeds

D. Finetuning

E. Additional results on model scaling behavior

F. Details on CodeContests finetuning

G. Additional results on natural language benchmarks

H. Additional results on abstractive text summarization

I. Additional results on mathematical reasoning in natural language

J. Additional results on induction learning

K. Additional results on algorithmic reasoning

L. Additional intuitions on multi-token prediction

M. Training hyperparameters

G. Additional results on natural language benchmarks

We evaluate the models from Section 3.7 on standard natural language processing benchmarks: ARC Challenge (Yadav et al., 2019), COPA (Roemmele et al., 2011), Hellaswag (Zellers et al., 2019), Natural Questions (Kwiatkowski et al., 2019), PIQA (Bisk et al., 2019), SIQA (Sap et al., 2019) and TriviaQA (Joshi et al., 2017).