Achilles' Heel of Mamba: Essential difficulties of the Mamba architecture demonstrated by synthetic data

NeurIPSSpotlight2025

Authors: Tianyi Chen, Pengxiao Lin, Zhiwei Wang, Zhi-Qin John Xu
Venue: NeurIPS 2025
Track: Spotlight

TL;DR

State Space Models (SSMs) have emerged as promising alternatives to attention mechanisms, with the Mamba architecture demonstrating impressive performance and linear complexity for processing long seq…

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

state space model attention mamba

← All NeurIPS 2025 Spotlight papers · Browse the whole archive