# Accelerating LLM Inference via Vector Index Based Output Embeddings

arXiv: https://arxiv.org/abs/2608.27460 · HTML: https://mlsanity.com/paper/2608.27460/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 18 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Builds on (most cited first, 14 of 14)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphs](https://mlsanity.com/paper/1603.09320/) (2016)
- [Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputs](https://mlsanity.com/paper/2407.01082/) (2024)
- [Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS)](https://mlsanity.com/paper/1405.5869/) (2014)
- [Efficient softmax approximation for GPUs](https://mlsanity.com/paper/1609.04309/) (2016)
- [CAGRA: Highly Parallel Graph Construction and Approximate Nearest Neighbor Search for GPUs](https://mlsanity.com/paper/2308.15136/) (2023)

## Abstract (arXiv)

Large output embedding matrices create a significant memory bandwidth bottleneck during autoregressive decoding, especially for compact LLMs with large multilingual vocabularies. We reformulate the output projection followed by top-k token selection as a maximum inner product search over token embeddings and replace the dense vocabulary projection with an HNSW-based vector index. The resulting output head retrieves only a small candidate set of high-scoring tokens and can be integrated into existing decoding pipelines by scattering retrieved logits into a sparse full-vocabulary tensor. On CPU inference with Gemma 3, Llama 3.2, and Qwen 3 models, our method substantially accelerates the output projection and improves end-to-end batch-size-one decoding throughput by up to 82% for Gemma 3 270M, while preserving generation quality under AlpacaEval evaluation. These results suggest approximate retrieval is a practical alternative to dense output projections in latency-sensitive small-batch decoding.