← 研踪 PaperTrailarXiv:2611.00999 · v2
ISBC: Interpretable Sparse Binary Coding of Pretrained Text Embeddings
首次提交 2026-10-31 22:37
版本更新 2026-11-03 16:24
雷达观测 2026-11-04 23:39
摘要
Pretrained text embeddings encode semantic structure in high-dimensional continuous spaces whose coordinates are difficult to interpret. We introduce Interpretable Sparse Binary Coding (ISBC), which learns a small set of binary dimensions from frozen embeddings without taxonomy labels or downstream outcomes. A trainable encoder and affine decoder combine reconstruction with controls on activity, factor utilization, dependence, and activation uncertainty. Individual bits and their conjunctions select overlapping document groups; complete activation profiles provide finer contexts for conditional comparisons. We evaluate these representations across corpora, backbones, and code configurations through reconstruction, geometry, predictive probes, and document inspection. Empirical cases identify both distinctions within existing categories and shared content across their boundaries. In a mixed-domain corpus, the most frequent profiles predominantly identify source collections, while selected within-source comparisons reveal sentiment and migration contrasts. Selected associations persist in additional documents encoded without refitting. In an engagement application, bits alone improve prediction of log-transformed outcomes over a constant reference, while their combinations identify within-event content contrasts associated with engagement. The results support using binary coordinates and their combinations for empirical interpretation, with complete profiles providing a fine, partly initialization-dependent description rather than a unique discrete taxonomy.
作者备注：updated version for smoother reading of the results
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-03 16:24 · 最新观测版本
ISBC: Interpretable Sparse Binary Coding of Pretrained Text Embeddings
Pretrained text embeddings encode semantic structure in high-dimensional continuous spaces whose coordinates are difficult to interpret. We introduce Interpretable Sparse Binary Coding (ISBC), which learns a small set of binary dimensions from frozen embeddings without taxonomy labels or downstream outcomes. A trainable encoder and affine decoder combine reconstruction with controls on activity, factor utilization, dependence, and activation uncertainty. Individual bits and their conjunctions select overlapping document groups; complete activation profiles provide finer contexts for conditional comparisons. We evaluate these representations across corpora, backbones, and code configurations through reconstruction, geometry, predictive probes, and document inspection. Empirical cases identify both distinctions within existing categories and shared content across their boundaries. In a mixed-domain corpus, the most frequent profiles predominantly identify source collections, while selected within-source comparisons reveal sentiment and migration contrasts. Selected associations persist in additional documents encoded without refitting. In an engagement application, bits alone improve prediction of log-transformed outcomes over a constant reference, while their combinations identify within-event content contrasts associated with engagement. The results support using binary coordinates and their combinations for empirical interpretation, with complete profiles providing a fine, partly initialization-dependent description rather than a unique discrete taxonomy.
查看此版本解读 →查看此版本原文 ↗