This story on HackerNoon has a decentralized backup on Sia.
Transaction ID: 8m782nWDhOteoFixG6JcgJioYznWduk1kcUrC8yNTmQ
Cover

LLM Performance Scaling: Multi-Token Prediction Across Model Sizes

Written by @largemodels | Published on 2025/6/10
TL;DR —
This table provides a detailed comparison of multi-token and next-token prediction performance on HumanEval and MBPP across a wide range of LLM sizes.

Abstract and 1. Introduction

2. Method

3. Experiments on real data

3.1. Benefits scale with model size and 3.2. Faster inference

3.3. Learning global patterns with multi-byte prediction and 3.4. Searching for the optimal n

3.5. Training for multiple epochs and 3.6. Finetuning multi-token predictors

3.7. Multi-token prediction on natural language

4. Ablations on synthetic data and 4.1. Induction capability

4.2. Algorithmic reasoning

5. Why does it work? Some speculation and 5.1. Lookahead reinforces choice points

5.2. Information-theoretic argument

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

E. Additional results on model scaling behavior

Table S7: Scaling model size Full results of scaling model size with n=1,2 and 4.

This paper is available on arxiv under CC BY 4.0 DEED license.

Authors:

(1) Fabian Gloeckle, FAIR at Meta, CERMICS Ecole des Ponts ParisTech, and contributed equally;

(2) Badr Youbi IdrissiFAIR at Meta, LISN Université Paris-Saclay, and contributed equally;

(3) Baptiste Rozière, FAIR at Meta;

(4) David Lopez-Paz, FAIR at Meta and his the last author;

(5) Gabriel Synnaeve, FAIR at Meta and the last author.

[story continues]

Written by
@largemodels
The Large-ness of Large Language Models (LLMs) ushered in a technological revolution. We dissect the research.
Topics and
tags
multi-token-prediction|next-token-prediction|llm-comparison|coding-ai|deep-learning-scalability|llm-scaling|coding-benchmarks|humaneval
This story on HackerNoon has a decentralized backup on Sia.
Transaction ID: 8m782nWDhOteoFixG6JcgJioYznWduk1kcUrC8yNTmQ