# SPIN: Sparsifying and Integrating Internal Neurons in Large Language Models for Text Classification Source: [SPIN: Sparsifying and Integrating Internal Neurons in Large Language Models for Text Classification]() Collection index: [llms.txt]() Data updated: 2026-01-04T04:37:34.891Z Authors: Difan Jiao, Yilun Liu, Zhenwei Tang, Daniel Matter, J\\"urgen Pfeffer, Ashton Anderson Published: 2023-11-27T16:28:20.000Z ## Abstract Among the many tasks that Large Language Models (LLMs) have revolutionized is text classification. Current text classification paradigms, however, rely solely on the output of the final layer in the LLM, with the rich information contained in internal neurons largely untapped. In this study, we present SPIN: a model-agnostic framework that sparsifies and integrates internal neurons of intermediate layers of LLMs for text classification. Specifically, SPIN sparsifies internal neurons by linear probing-based salient neuron selection layer by layer, avoiding noise from unrelated neurons and ensuring efficiency. The cross-layer salient neurons are then integrated to serve as multi-layered features for the classification head. Extensive experimental results show our proposed SPIN significantly improves text classification accuracy, efficiency, and interpretability. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Text Classification on IMDB](): Text Classification; IMDB; Accuracy: 96.06. - [Text Classification on EDOS (test)](): Text Classification; EDOS (test); Macro F1 Score: 81.48. - [Text Classification on SST-2 (test)](): Text Classification; SST-2 (test); Accuracy: 95.64. ## Original sources and implementation [Original paper]() [PDF]() [Implementation]()