# Long-Context Demonstration Selection Using State Space Models

arXiv: https://arxiv.org/abs/2609.17888 · HTML: https://mlsanity.com/paper/2609.17888/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 15 Sep 2026.

## Builds on (most cited first, 24 of 24)

- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [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)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [What Can Transformers Learn In-Context? A Case Study of Simple Function Classes](https://mlsanity.com/paper/2208.01066/) (2022)
- [HiPPO: Recurrent Memory with Optimal Polynomial Projections](https://mlsanity.com/paper/2008.07669/) (2020)
- [Parallelizing Linear Transformers with the Delta Rule over Sequence Length](https://mlsanity.com/paper/2406.06484/) (2024)
- [Learning to Compress Prompts with Gist Tokens](https://mlsanity.com/paper/2304.08467/) (2023)
- [Adapting Language Models to Compress Contexts](https://mlsanity.com/paper/2305.14788/) (2023)
- [DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads](https://mlsanity.com/paper/2410.10819/) (2024)
- [Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling](https://mlsanity.com/paper/2406.07522/) (2024)
- [PaTH Attention: Position Encoding via Accumulating Householder Transformations](https://mlsanity.com/paper/2505.16381/) (2025)
- [Precise High-Dimensional Asymptotics for Quantifying Heterogeneous Transfers](https://mlsanity.com/paper/2010.11750/) (2020)
- [From Few to Many: Self-Improving Many-Shot Reasoners Through Iterative Optimization and Generation](https://mlsanity.com/paper/2502.00330/) (2025)
- [Scalable Multitask Learning Using Gradient-based Estimation of Task Affinity](https://mlsanity.com/paper/2409.06091/) (2024)
- [Boosting Multitask Learning on Graphs through Higher-Order Task Affinities](https://mlsanity.com/paper/2306.14009/) (2023)
- [Scalable Fine-tuning from Multiple Data Sources: A First-Order Approximation Approach](https://mlsanity.com/paper/2409.19458/) (2024)
- [Data-Driven Switchback Experiments: Theoretical Tradeoffs and Empirical Bayes Designs](https://mlsanity.com/paper/2406.06768/) (2024)
- [Efficient Many-Shot In-Context Learning with Dynamic Block-Sparse Attention](https://mlsanity.com/paper/2503.08640/) (2025)
- [Efficient Ensemble for Fine-tuning Language Models on Multiple Datasets](https://mlsanity.com/paper/2505.21930/) (2025)
- [Linear-Time Demonstration Selection for In-Context Learning via Gradient Estimation](https://mlsanity.com/paper/2508.19999/) (2025)
- [Scalable Multi-Objective and Meta Reinforcement Learning via Gradient Estimation](https://mlsanity.com/paper/2511.12779/) (2025)
- [Efficient Estimation of Kernel Surrogate Models for Task Attribution](https://mlsanity.com/paper/2602.03783/) (2026)
- [Flash Inference: Near Linear Time Inference for Long Convolution Sequence Models and Beyond](https://mlsanity.com/paper/2410.12982/) (2024)

## Abstract (arXiv)

We study the problem of demonstration selection, which involves selecting a subset of examples for prepending to a query to a language model. This problem is closely related to in-context learning and language model inference. Since the inference cost of a transformer model scales quadratically with sequence length, the selection problem becomes especially challenging in a long-context scenario. In this paper, we tackle this problem by building on state space models (SSMs), which require only linear inference time given the input. Our approach involves two algorithms. The first learns a small set of SSMs through distillation of a (trained) transformer model. We partition all the layers into consecutive groups. Then for each group, we estimate a separate state space model to replicate the input-output behavior within the adjacent layers. Second, we map the distilled model outputs to a small set of tokens, and apply these embeddings for demonstration selection in downstream applications. We perform extensive experiments in both synthetic and real-world datasets to validate our approach. We demonstrate that the distilled SSMs only incur an approximation error of less than 0.7% relative to the true output. In downstream evaluation, we show that on several text classification and reasoning tasks, our approach reduces FLOPs by 14.2× and improves accuracy by 6.48% relative to baseline demonstration selection methods.