# Direct Alignment of Draft Model for Speculative Decoding with Chat-Fine-Tuned LLMs

arXiv: https://arxiv.org/abs/2403.00858 · HTML: https://mlsanity.com/paper/2403.00858/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving.
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 29 Feb 2024.

## Placed in

- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

## Builds on (most cited first, 20 of 20)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [High-Dimensional Continuous Control Using Generalized Advantage Estimation](https://mlsanity.com/paper/1506.02438/) (2015)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond](https://mlsanity.com/paper/1602.06023/) (2016)
- [Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/1606.07947/) (2016)
- [The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Only](https://mlsanity.com/paper/2306.01116/) (2023)
- [OpenAssistant Conversations -- Democratizing Large Language Model Alignment](https://mlsanity.com/paper/2304.07327/) (2023)
- [Baichuan 2: Open Large-scale Language Models](https://mlsanity.com/paper/2309.10305/) (2023)
- [DistillSpec: Improving Speculative Decoding via Knowledge Distillation](https://mlsanity.com/paper/2310.08461/) (2023)
- [f-Divergence Minimization for Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/2307.15190/) (2023)
- [On Reinforcement Learning and Distribution Matching for Fine-Tuning Language Models with no Catastrophic Forgetting](https://mlsanity.com/paper/2206.00761/) (2022)
- [Autoregressive Knowledge Distillation through Imitation Learning](https://mlsanity.com/paper/2009.07253/) (2020)
- [Weight subcloning: direct initialization of transformers using larger pretrained ones](https://mlsanity.com/paper/2312.09299/) (2023)

## Built on by (most cited first, 22 of 22)

- [KeyDiff: Key Similarity-Based KV Cache Eviction for Long-Context LLM Inference in Resource-Constrained Environments](https://mlsanity.com/paper/2504.15364/) (2025)
- [On Speculative Decoding for Multimodal Large Language Models](https://mlsanity.com/paper/2404.08856/) (2024)
- [AdaEDL: Early Draft Stopping for Speculative Decoding of Large Language Models via an Entropy-based Lower Bound on Token Acceptance Probability](https://mlsanity.com/paper/2410.18351/) (2024)
- [Structuring The Future: Diffusion LLM Speculative Decoding via Calibrated Draft Graphs](https://mlsanity.com/paper/2509.18085/) (2025)
- [JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting](https://mlsanity.com/paper/2606.18394/) (2026)
- [VOCABTRIM: Vocabulary Pruning for Efficient Speculative Decoding in LLMs](https://mlsanity.com/paper/2506.22694/) (2025)
- [DEL: Context-Aware Dynamic Exit Layer for Efficient Self-Speculative Decoding](https://mlsanity.com/paper/2504.05598/) (2025)
- [Closer Look at Efficient Inference Methods: A Survey of Speculative Decoding](https://mlsanity.com/paper/2411.13157/) (2024)
- [FastDraft: How to Train Your Draft](https://mlsanity.com/paper/2411.11055/) (2024)
- [Advancing Decoding Strategies: Enhancements in Locally Typical Sampling for LLMs](https://mlsanity.com/paper/2506.05387/) (2025)
- [DIVERSED: Relaxed Speculative Decoding via Dynamic Ensemble Verification](https://mlsanity.com/paper/2604.07622/) (2026)
- [Mirror Speculative Decoding: Breaking the Serial Barrier in LLM Inference](https://mlsanity.com/paper/2510.13161/) (2025)
- [Adversarial Prompts for Acceptance Collapse in Speculative Decoding](https://mlsanity.com/paper/2607.21804/) (2026)
- [ConFu: Contemplate the Future for Better Speculative Sampling](https://mlsanity.com/paper/2603.08899/) (2026)
- [Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing](https://mlsanity.com/paper/2603.17942/) (2026)
- [ELMoE-3D: Leveraging Intrinsic Elasticity of MoE for Hybrid-Bonding-Enabled Self-Speculative Decoding in On-Premises Serving](https://mlsanity.com/paper/2604.14626/) (2026)
- [Speculative Decoding and the Curse of Multilinguality](https://mlsanity.com/paper/2605.30580/) (2026)
- [Speculative Decoding at Temperature Zero: A Scoped Safety-Invariance Screen with a 48,072-Sample Expansion](https://mlsanity.com/paper/2606.25097/) (2026)
- [Recovering Off-Policy Supervision for Speculative Decoding](https://mlsanity.com/paper/2609.38795/) (2026)
- [CAST: Cost-Aware Speculative Trees from One-Pass Block Drafters](https://mlsanity.com/paper/2610.00321/) (2026)
- [How Divergence Becomes Decision Flips in Compressed Language Models](https://mlsanity.com/paper/2610.00694/) (2026)
- [JetSpec++: Foresight Parallel Drafting with Self Refinements](https://mlsanity.com/paper/2610.19792/) (2026)

## Abstract (arXiv)

Text generation with Large Language Models (LLMs) is known to be memory bound due to the combination of their auto-regressive nature, huge parameter counts, and limited memory bandwidths, often resulting in low token rates. Speculative decoding has been proposed as a solution for LLM inference acceleration. However, since draft models are often unavailable in the modern open-source LLM families, e.g., for Llama 2 7B, training a high-quality draft model is required to enable inference acceleration via speculative decoding. In this paper, we propose a simple draft model training framework for direct alignment to chat-capable target models. With the proposed framework, we train Llama 2 Chat Drafter 115M, a draft model for Llama 2 Chat 7B or larger, with only 1.64% of the original size. Our training framework only consists of pretraining, distillation dataset generation, and finetuning with knowledge distillation, with no additional alignment procedure. For the finetuning step, we use instruction-response pairs generated by target model for distillation in plausible data distribution, and propose a new Total Variation Distance++ (TVD++) loss that incorporates variance reduction techniques inspired from the policy gradient method in reinforcement learning. Our empirical results show that Llama 2 Chat Drafter 115M with speculative decoding achieves up to 2.3 block efficiency and 2.4× speed-up relative to autoregressive decoding on various tasks with no further task-specific fine-tuning.