arXiv:2607. 25546v1 Announce Type: new Abstract: Given a model that is already trained, which features does it rely on causally versus spuriously?
arXiv:2609.36881v1 Announce Type: new Abstract: Causal foundation models (CFMs) pre-trained on data generated from various structural causal models (SCMs) have been proposed for estimating causal eff...
arXiv:2607. 19618v1 Announce Type: cross Abstract: Genomic language models achieve strong performance across regulatory-genomics tasks, yet what these models internally represent remains opaque, and the field lacks a principled procedure for verifying that an apparent ``concept'' inside a model is real rather than an artifact of sequence composition.
arXiv:2609.36771v1 Announce Type: cross Abstract: Root cause analysis (RCA) is a critical problem in many real-world scenarios. RCA enables the identification of faulty or failing mechanisms in a sys...
arXiv:2607. 25532v1 Announce Type: new Abstract: Consider a model trained at a single hospital to predict patient recovery, where the measured feature $X$ bundles the patient's true health signal ($C$) with a systematic artefact from that hospital's equipment ($S$).
arXiv:2605. 09169v2 Announce Type: replace-cross Abstract: A Mamba state-space model trained only for next-step prediction appears to recover Granger-causal structure through a simple readout $S = |W_{out} W_{in}|$, with early experiments suggesting the phenomenon generalized across architectures and benefited from interventional data at $p < 10^{-5}$.
arXiv:2607. 20596v1 Announce Type: new Abstract: Sparse autoencoder (SAE) features are used to interpret and steer large language models, yet whether a feature's causal role is stable across SAE families remains untested.
arXiv:2609.31315v1 Announce Type: cross Abstract: Unobserved common causes are pervasive in real-world time series and can induce spurious associations that causal discovery methods mistake for direc...
The study evaluates 12 instruction‑tuned open‑weight LLMs on six causal‑graph benchmarks, testing five prompting strategies and four confidence sources. Findings show that LLMs tend to over‑predict edges, misclassify indirect or reversed edges as direct, and exhibit high over‑confidence, while conventional confidence estimates are unreliable and agreement signals offer limited improvement. The results suggest LLMs should be used as externally validated soft causal priors rather than definitive causal‑structure evidence.
arXiv:2606. 17113v1 Announce Type: new Abstract: Distinguishing causal adverse drug events (ADEs) from spurious correlations remains a central challenge in pharmacovigilance.
Linear probes can decode safety‑relevant concepts such as truthfulness from language‑model activations, but probe accuracy may reflect only decodability, not causal influence on model behavior. The authors show that probe weight geometry alone cannot identify the features the model actually uses, because geometrically aligned features need not be causally relevant. They introduce a sparse‑autoencoder (SAE) decomposition that ranks features by probe alignment and gradient sensitivity, and demonstrate that ablating shared, probe‑only, and random feature sets reveals a sharp dissociation: shared features drive model output changes far more than probe‑only or random features, confirming that causal relevance requires intervention beyond weight geometry.
arXiv:2609.37664v1 Announce Type: new Abstract: Causal Normalizing Flows (CNFs) enable causal inference from observational data given the causal structure, but they assume fully observed training dat...
The paper introduces a model‑agnostic inference framework for partially identified causal effects that leverages covariate information without requiring discrete covariates or accurate conditional distribution estimates. Using duality theory for optimal transport, the method delivers uniformly valid inference in randomized experiments, is doubly robust in observational settings, achieves asymptotic unbiasedness when nuisance parameters converge semiparametrically, and allows multiplier‑bootstrap selection of covariates and models while remaining computationally efficient. Empirical applications show the approach consistently narrows identified sets and confidence intervals without imposing extra structural assumptions.