Metraipolis › archive › dc_fd12f901dd47

Exploring Sparse Autoencoders in Text-Based Causal Confounding Adjustment — arXiv:2609.01322

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01322
License (as archived)
CC0-1.0
Archived
2026-10-17T06:41:34.260Z
SHA-256 of the archived text
1dd95b54994897b96d20bca74c7c3182108c4d564fd437915bfa54f9ad93c286
Stable URL of this page
https://metraipolis.com/documento/dc_fd12f901dd47

How to cite

Exploring Sparse Autoencoders in Text-Based Causal Confounding Adjustment — arXiv:2609.01322 — arXiv (en), https://arxiv.org/abs/2609.01322 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fd12f901dd47 (Metraipolis archive dc_fd12f901dd47, sha256 1dd95b54994897b96d20bca74c7c3182108c4d564fd437915bfa54f9ad93c286)
BibTeX
@misc{dcfd12f901dd47,
  title = {Exploring Sparse Autoencoders in Text-Based Causal Confounding Adjustment — arXiv:2609.01322},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01322},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fd12f901dd47; sha256 1dd95b54994897b96d20bca74c7c3182108c4d564fd437915bfa54f9ad93c286},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01322. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In many settings, studying causal questions based on text data requires adjusting for confounding information within texts. Yet there is a tradeoff in constructing text representations for adjustment: they must be sufficiently large and/or dense to preserve the confounding variables necessary for unbiased effect estimation, but sufficiently small and/or sparse to satisfy finite-sample overlap and yield low-variance estimates. To address this tradeoff, we turn to sparse autoencoders (SAEs), and propose a novel causal adjustment pipeline that iteratively selects a minimal set of SAE features via conditional independence tests. We find that SAE representations achieve better adjustments (lower bias and and higher coverage) than alternative representations in standard semi-synthetic evaluations with binary confounders, and their interpretability offers opportunities for falsification. We also introduce a more realistic semi-synthetic evaluation that uses multi-label data as the unobserved confounders and find off-the-shelf adjustment methods require increased investigation for these more complex settings. Code: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.01322 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T06:33:38.855Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).