# Stability Implies Redundancy: Delta Attention Selective Halting for Efficient Long-Context Prefilling

arXiv: https://arxiv.org/abs/2604.18103 · HTML: https://mlsanity.com/paper/2604.18103/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 19 of its 32 references that are in the map.
- First posted on arXiv 20 Apr 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)

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

- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression](https://mlsanity.com/paper/2403.12968/) (2024)
- [dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching](https://mlsanity.com/paper/2506.06295/) (2025)
- [Keyformer: KV Cache Reduction through Key Tokens Selection for Efficient Generative Inference](https://mlsanity.com/paper/2403.09054/) (2024)
- [What Matters in Transformers? Not All Attention is Needed](https://mlsanity.com/paper/2406.15786/) (2024)
- [Joint Token Pruning and Squeezing Towards More Aggressive Compression of Vision Transformers](https://mlsanity.com/paper/2304.10716/) (2023)
- [Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters](https://mlsanity.com/paper/2406.12335/) (2024)
- [LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference](https://mlsanity.com/paper/2407.14057/) (2024)
- [Dynamic Token Pruning in Plain Vision Transformers for Semantic Segmentation](https://mlsanity.com/paper/2308.01045/) (2023)
- [A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Task](https://mlsanity.com/paper/2402.11917/) (2024)
- [Filter, Correlate, Compress: Training-Free Token Reduction for MLLM Acceleration](https://mlsanity.com/paper/2411.17686/) (2024)
- [SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning](https://mlsanity.com/paper/2508.06447/) (2025)
- [Mechanism and Emergence of Stacked Attention Heads in Multi-Layer Transformers](https://mlsanity.com/paper/2411.12118/) (2024)
- [GraphKV: Breaking the Static Selection Paradigm with Graph-Based KV Cache Eviction](https://mlsanity.com/paper/2509.00388/) (2025)

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

- [STaR-KV: Spatio-Temporal Adaptive Re-weighting for KV Cache Compression in GUI Vision-Language Models](https://mlsanity.com/paper/2606.01790/) (2026)

## Abstract (arXiv)

Prefilling computational costs pose a significant bottleneck for Large Language Models (LLMs) and Large Multimodal Models (LMMs) in long-context settings. While token pruning reduces sequence length, prior methods rely on heuristics that break compatibility with hardware-efficient kernels like FlashAttention. In this work, we observe that tokens evolve toward textitsemantic fixing points, making further processing redundant. To this end, we introduce Delta Attention Selective Halting (DASH), a training-free policy that monitors the layer-wise update dynamics of the self-attention mechanism to selectively halt stabilized tokens. Extensive evaluation confirms that DASH generalizes across language and vision benchmarks, delivering significant prefill speedups while preserving model accuracy and hardware efficiency. Code will be released at https://github.com/verach3n/DASH.git.