# Intra-Prompt Parallel Decoding for Common-Context Question Answering

arXiv: https://arxiv.org/abs/2609.05707 · HTML: https://mlsanity.com/paper/2609.05707/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Prompt and context compression, one of 12 directions in Efficient LLM inference and serving.
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 4 Sep 2026.

## Placed in

- Efficient LLM inference and serving / [Prompt and context compression](https://mlsanity.com/inference/prompt-and-context-compression/)

## Builds on (most cited first, 19 of 19)

- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [RACE: Large-scale ReAding Comprehension Dataset From Examinations](https://mlsanity.com/paper/1704.04683/) (2017)
- [Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads](https://mlsanity.com/paper/2401.10774/) (2024)
- [The NarrativeQA Reading Comprehension Challenge](https://mlsanity.com/paper/1712.07040/) (2017)
- [Break the Sequential Dependency of LLM Inference Using Lookahead Decoding](https://mlsanity.com/paper/2402.02057/) (2024)
- [A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systems](https://mlsanity.com/paper/2402.18013/) (2024)
- [Skeleton-of-Thought: Prompting LLMs for Efficient Parallel Generation](https://mlsanity.com/paper/2307.15337/) (2023)
- [A Comprehensive Survey on Process-Oriented Automatic Text Summarization with Exploration of LLM-Based Methods](https://mlsanity.com/paper/2403.02901/) (2024)
- [ChunkAttention: Efficient Self-Attention with Prefix-Aware KV Cache and Two-Phase Partition](https://mlsanity.com/paper/2402.15220/) (2024)
- [Hydragen: High-Throughput LLM Inference with Shared Prefixes](https://mlsanity.com/paper/2402.05099/) (2024)
- [A Survey of Large Language Model Agents for Question Answering](https://mlsanity.com/paper/2503.19213/) (2025)
- [Mind the Memory Gap: Unveiling GPU Bottlenecks in Large-Batch LLM Inference](https://mlsanity.com/paper/2503.08311/) (2025)
- [Marconi: Prefix Caching for the Era of Hybrid LLMs](https://mlsanity.com/paper/2411.19379/) (2024)
- [APAR: LLMs Can Do Auto-Parallel Auto-Regressive Decoding](https://mlsanity.com/paper/2401.06761/) (2024)
- [ExtractGPT: Exploring the Potential of Large Language Models for Product Attribute Value Extraction](https://mlsanity.com/paper/2310.12537/) (2023)
- [DeFT: Decoding with Flash Tree-attention for Efficient Tree-structured LLM Inference](https://mlsanity.com/paper/2404.00242/) (2024)

## Built on by (most cited first, 1 of 1)

- [Scaling E-Commerce Attribute Extraction with Parallel Decoding](https://mlsanity.com/paper/2609.09716/) (2026)

## Abstract (arXiv)

In common-context question answering (CCQA) tasks, multiple input questions share a common context to base their answers from. However, Large Language Models typically generate each answer using an independent prompt. While existing batching and caching techniques help improve parallelism and reduce repeated computations, the separation of questions across prompts limits the achievable speedup, as modern GPUs are underutilized due to a memory bottleneck during attention. We present Intra-Prompt Parallel Decoding (IPPD), a novel inference method that answers multiple common-context questions in parallel within a single prompt. IPPD directly addresses the bottleneck by efficiently sharing both memory and computation during the attention process, as the next token for every question is decoded in a single inference step. IPPD uses virtual position IDs and attention mask manipulation to generate the same output as standard prompting without requiring fine-tuning or any changes to the LLM architecture. Since all parallelism occurs within a prompt, IPPD is fully compatible with batched inference, even when each prompt features a different context. Our experiments show that IPPD delivers up to 7X the effective throughput as standard decoding without quality degradation, and outperforms prefix caching with PagedAttention in most settings.