Evidence-Type Competition: When Can Interventional Data Teach Language Models Causal Direction?

• URL: http://arxiv.org/abs/2607.29484v1
• Date: Fri, 31 Jul 2026 14:49:18 GMT
• Title: Evidence-Type Competition: When Can Interventional Data Teach Language Models Causal Direction?
• Abstract summary: interventional data is widely regarded as the gold standard for teaching models causal reasoning.<n>We test this assumption in a fully controlled synthetic environment pitting observational correlation against causal effect.<n>What governs whether interventional evidence is used is not the training mixture but the evidence type present in the context at inference time.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Interventional data is widely regarded as the gold standard for teaching models causal reasoning. We test this assumption in a fully controlled synthetic environment pitting observational correlation against causal effect, and find it fails instructively. In Simpson's-paradox worlds, where the two have systematically opposite signs, increasing the fraction of interventional samples in pretraining does not improve causal direction: the magnitude of the model's do()-response grows monotonically, yet its sign is copied from the observational context. What governs whether interventional evidence is used is not the training mixture but the evidence type present in the context at inference time. Under an identical training recipe, a purely observational context induces systematic sign reversal in 29/50 worlds, a mixed context in 19/50, while aligned interventional probes alone yield 41/50 correct. Erasing observational evidence from the context immediately releases the suppressed causal interpolation ability (ratio_true = +0.56); a four-state content manipulation shows the switch is content-mediated and graded. The suppression is stable across training seeds (11/11 strong reversals persist on a matched-protocol second seed) and robust as a rate at 0.93B parameters (31.8% vs. 6% reversals in the matched probe-only arm), even as absolute gains shrink four-fold. An external audit on CLadder exposes a learned positive-effect prior with a two-layer structure: sign-randomized retraining removes it in-distribution but not out-of-distribution. We summarize: the capability lives in the weights; the switch lives in the context, and activation patching localizes the switch to the middle layers' observational rows. We further quantify the sampling noise floor of probe-based causal evaluation and an evidence-averaging protocol that cuts sign errors from 26% to 9%.

Related papers

• When Outputs Disperse, Does Epistemic Revision Follow? A Black-Box Diagnostic for Machine Collectives [0.0]
  We operationalize dispersion-revision coupling in collective intelligence research.<n>We show that conditional dissent improves false-premise recovery by +17.7 points.<n>We recommend reporting per-intervention stance shift and premise-preservation rate alongside accuracy.
  arXiv  Detail & Related papers  (2026-08-04T14:19:59Z)
• Beyond Tracking or Shortcut: Composition-Bounded Predictive States in Poker Autoregressive Models [4.810191717262418]
  Hidden-state probes often recover latent labels in imperfect-information sequence models.<n>This paper studies this ambiguity in a no-range Limit Hold'em autoregressive model trained only on action and value targets.
  arXiv  Detail & Related papers  (2026-06-13T15:56:26Z)
• Hallucination as Trajectory Commitment: Causal Evidence for Asymmetric Attractor Dynamics in Transformer Generation [0.0]
  hallucination in autoregressive language models is governed by asymmetric attractor dynamics.<n>We use same-prompt bifurcation to isolate trajectory dynamics from prompt-level confounds.
  arXiv  Detail & Related papers  (2026-04-16T12:16:53Z)
• Evaluating Evidence Grounding Under User Pressure in Instruction-Tuned Language Models [6.431677598656395]
  In contested domains, instruction-tuned language models must balance user-alignment pressures against faithfulness to the in-context evidence.<n>We conduct fine-grained ablations over evidence composition and uncertainty cues across 19 instruction-tuned models spanning 0.27B to 32B parameters.
  arXiv  Detail & Related papers  (2026-03-20T17:38:23Z)
• Before the First Token: Scale-Dependent Emergence of Hallucination Signals in Autoregressive Language Models [6.5607014897534865]
  We study the temporal dynamics of hallucination-indicative internal representations across 7 autoregressive transformers.<n>We identify a scale-dependent phase transition: models below 400M parameters show chance-level probe accuracy at every generation position.<n>At the 7B scale, we observe a striking dissociation: Pythia-6.9B produces a flat temporal profile, while instruction-tuned Qwen2.5-7B shows a dominant pre-generation effect.
  arXiv  Detail & Related papers  (2026-03-20T02:30:01Z)
• Causal Understanding by LLMs: The Role of Uncertainty [43.87879175532034]
  Recent papers show LLMs achieve near-random accuracy in causal relation classification.<n>We investigate whether pretraining exposure to causal examples improves causal understanding.
  arXiv  Detail & Related papers  (2025-09-24T13:06:35Z)
• BURN: Backdoor Unlearning via Adversarial Boundary Analysis [73.14147934175604]
  Backdoor unlearning aims to remove backdoor-related information while preserving the model's original functionality.<n>We propose Backdoor Unlearning via adversaRial bouNdary analysis (BURN), a novel defense framework that integrates false correlation decoupling, progressive data refinement, and model purification.
  arXiv  Detail & Related papers  (2025-07-14T17:13:06Z)
• Counterfactual Realizability [52.85109506684737]
  We introduce a formal definition of realizability, the ability to draw samples from a distribution, and then develop a complete algorithm to determine whether an arbitrary counterfactual distribution is realizable.<n>We illustrate the implications of this new framework for counterfactual data collection using motivating examples from causal fairness and causal reinforcement learning.
  arXiv  Detail & Related papers  (2025-03-14T20:54:27Z)
• Can Active Sampling Reduce Causal Confusion in Offline Reinforcement Learning? [58.942118128503104]
  Causal confusion is a phenomenon where an agent learns a policy that reflects imperfect spurious correlations in the data. This phenomenon is particularly pronounced in domains such as robotics. In this paper, we study causal confusion in offline reinforcement learning.
  arXiv  Detail & Related papers  (2023-12-28T17:54:56Z)
• Nonparametric Identifiability of Causal Representations from Unknown Interventions [63.1354734978244]
  We study causal representation learning, the task of inferring latent causal variables and their causal relations from mixtures of the variables. Our goal is to identify both the ground truth latents and their causal graph up to a set of ambiguities which we show to be irresolvable from interventional data.
  arXiv  Detail & Related papers  (2023-06-01T10:51:58Z)
• Adversarial Robustness through the Lens of Causality [105.51753064807014]
  adversarial vulnerability of deep neural networks has attracted significant attention in machine learning. We propose to incorporate causality into mitigating adversarial vulnerability. Our method can be seen as the first attempt to leverage causality for mitigating adversarial vulnerability.
  arXiv  Detail & Related papers  (2021-06-11T06:55:02Z)
• On Disentangled Representations Learned From Correlated Data [59.41587388303554]
  We bridge the gap to real-world scenarios by analyzing the behavior of the most prominent disentanglement approaches on correlated data. We show that systematically induced correlations in the dataset are being learned and reflected in the latent representations. We also demonstrate how to resolve these latent correlations, either using weak supervision during training or by post-hoc correcting a pre-trained model with a small number of labels.
  arXiv  Detail & Related papers  (2020-06-14T12:47:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.