# LLM as Effective Streaming Processor: Bridging Streaming-Batch Mismatches with Group Position Encoding

arXiv: https://arxiv.org/abs/2505.16983 · HTML: https://mlsanity.com/paper/2505.16983/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 36 of its 49 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 30 of 36)

- [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)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation](https://mlsanity.com/paper/2108.12409/) (2021)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities](https://mlsanity.com/paper/2305.11000/) (2023)
- [Large Language Models Are State-of-the-Art Evaluators of Translation Quality](https://mlsanity.com/paper/2302.14520/) (2023)
- [The Impact of Positional Encoding on Length Generalization in Transformers](https://mlsanity.com/paper/2305.19466/) (2023)
- [AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling](https://mlsanity.com/paper/2402.12226/) (2024)
- [Transformer Dissection: A Unified Understanding of Transformer's Attention via the Lens of Kernel](https://mlsanity.com/paper/1908.11775/) (2019)
- [Transformer Language Models without Positional Encodings Still Learn Positional Information](https://mlsanity.com/paper/2203.16634/) (2022)
- [Randomized Positional Encodings Boost Length Generalization of Transformers](https://mlsanity.com/paper/2305.16843/) (2023)
- [DiSAN: Directional Self-Attention Network for RNN/CNN-Free Language Understanding](https://mlsanity.com/paper/1709.04696/) (2017)
- [CAPE: Encoding Relative Positions with Continuous Augmented Positional Embeddings](https://mlsanity.com/paper/2106.03143/) (2021)
- [TransLLaMa: LLM-based Simultaneous Translation System](https://mlsanity.com/paper/2402.04636/) (2024)
- [Direct Simultaneous Speech-to-Text Translation Assisted by Synchronized Streaming ASR](https://mlsanity.com/paper/2106.06636/) (2021)
- [Streaming Simultaneous Speech Translation with Augmented Memory Transformer](https://mlsanity.com/paper/2011.00033/) (2020)
- [Learning When to Translate for Streaming Speech](https://mlsanity.com/paper/2109.07368/) (2021)
- [Hidden Markov Transformer for Simultaneous Machine Translation](https://mlsanity.com/paper/2303.00257/) (2023)
- [SiLLM: Large Language Models for Simultaneous Machine Translation](https://mlsanity.com/paper/2402.13036/) (2024)
- [Non-autoregressive Streaming Transformer for Simultaneous Translation](https://mlsanity.com/paper/2310.14883/) (2023)

## Built on by (most cited first, 9 of 9)

- [HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early Exit](https://mlsanity.com/paper/2602.23699/) (2026)
- [VisiPruner: Decoding Discontinuous Cross-Modal Dynamics for Efficient Multimodal LLMs](https://mlsanity.com/paper/2510.17205/) (2025)
- [StreamingThinker: Large Language Models Can Think While Reading](https://mlsanity.com/paper/2510.17238/) (2025)
- [Speak While Watching: Unleashing TRUE Real-Time Video Understanding Capability of Multimodal Large Language Models](https://mlsanity.com/paper/2601.06843/) (2026)
- [What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language Models](https://mlsanity.com/paper/2603.00510/) (2026)
- [Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models](https://mlsanity.com/paper/2603.02872/) (2026)
- [From Static Inference to Dynamic Interaction: A Survey of Streaming Large Language Models](https://mlsanity.com/paper/2603.04592/) (2026)
- [ProactiveLLM: Learning Active Interaction for Streaming Large Language Models](https://mlsanity.com/paper/2606.00523/) (2026)
- [AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization](https://mlsanity.com/paper/2606.14694/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are primarily designed for batch processing. Existing methods for adapting LLMs to streaming rely either on expensive re-encoding or specialized architectures with limited scalability. This work identifies three key mismatches in adapting batch-oriented LLMs to streaming: (1) input-attention, (2) output-attention, and (3) position-ID mismatches. While it is commonly assumed that the latter two mismatches require frequent re-encoding, our analysis reveals that only the input-attention mismatch significantly impacts performance, indicating re-encoding outputs is largely unnecessary. To better understand this discrepancy with the common assumption, we provide the first comprehensive analysis of the impact of position encoding on LLMs in streaming, showing that preserving relative positions within source and target contexts is more critical than maintaining absolute order. Motivated by the above analysis, we introduce a group position encoding paradigm built on batch architectures to enhance consistency between streaming and batch modes. Extensive experiments on cross-lingual and cross-modal tasks demonstrate that our method outperforms existing approaches. Our method requires no architectural modifications, exhibits strong generalization in both streaming and batch modes. The code is available at repository https://github.com/EIT-NLP/StreamingLLM.