Sparsified State-Space Models are Efficient Highway Networks
Abstract
Simba, a hierarchical sparsification method for state-space models, enhances efficiency and information flow in natural language tasks by pruning tokens more aggressively in upper layers.
State-space models (SSMs) offer a promising architecture for sequence modeling, providing an alternative to Transformers by replacing expensive self-attention with linear recurrences. In this paper, we propose a simple yet effective trick to enhance SSMs within given computational budgets by sparsifying them. Our intuition is that tokens in SSMs are highly redundant due to gradual recurrent updates, and dense recurrence operations block the delivery of past information. In particular, we observe that upper layers of SSMs tend to be more redundant as they encode global information, while lower layers encode local information. Motivated by this, we introduce Simba, a hierarchical sparsification method for SSMs based on token pruning. Simba sparsifies upper layers more than lower layers, encouraging the upper layers to behave like highways. To achieve this, we propose a novel token pruning criterion for SSMs, measuring the global impact of tokens on the final output by accumulating local recurrences. We demonstrate that Simba outperforms the baseline model, Mamba, with the same FLOPS in various natural language tasks. Moreover, we illustrate the effect of highways, showing that Simba not only enhances efficiency but also improves the information flow across long sequences. Code is available at https://github.com/woominsong/Simba.
Community
In this paper, we propose a simple yet effective trick to enhance SSMs within given computational budgets using token pruning. Simba sparsifies upper layers more than lower layers, encouraging the upper layers to behave like highways. To achieve this, we propose a novel token pruning criterion for SSMs, measuring the global impact of tokens on the final output by accumulating local recurrences.
We demonstrate that Simba outperforms the baseline model, Mamba, with the same FLOPS in various natural language tasks. Moreover, we illustrate the effect of highways, showing that Simba not only enhances efficiency but also improves the information flow across long sequences.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Efficient Unstructured Pruning of Mamba State-Space Models for Resource-Constrained Environments (2025)
- LongMamba: Enhancing Mamba's Long-Context Capabilities via Training-Free Receptive Field Enlargement (2025)
- Balancing Computation Load and Representation Expressivity in Parallel Hybrid Neural Networks (2025)
- Compress, Gather, and Recompute: REFORMing Long-Context Processing in Transformers (2025)
- Leave it to the Specialist: Repair Sparse LLMs with Sparse Fine-Tuning via Sparsity Evolution (2025)
- Random Long-Context Access for Mamba via Hardware-aligned Hierarchical Sparse Attention (2025)
- Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper