Tag
A preprint presenting a sparse-recovery decoder that localizes up to 64 stuck cells in a 256×256 simulated analog crossbar using only 4–12 probe reads instead of 256, along with a 16-parameter drift calibration, while transparently reporting several pre-registered claims that failed.
This paper identifies sufficient sample sizes for maximum-likelihood support recovery of sparse binary signals in high-SNR regimes, revealing an information-theoretic threshold and trade-offs between measurement sparsity and computational cost.
STRIDE is a new framework for training data attribution in LLMs that models functional effects in activation space using sparse recovery and steering operators, achieving state-of-the-art accuracy with 13x speedup over previous methods.