Wrecking-ball interventions that collapse global model performance are empirically identified in EEG foundation models.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• wrecking-ball interventions that collapse global model performancequote0.911
  Load-bearing phrase describing catastrophic steering effects.
• Concept interventions on some concepts act as 'wrecking-ball' interventions, collapsing global model performance.finding0.861
  Observation of catastrophic performance drop when steering certain concepts.
• Can concept steering interventions on EEG foundation models be made selective rather than globally destructive?question0.776
  Research question motivating the introduction of the probe area metric and identification of operational regimes
• EEG foundation models achieve state-of-the-art clinical performance yet their internal computations remain opaque, constituting a barrier to clinical trust.claim0.763
  Motivating claim for the entire paper
• Out-of-distribution collapse in foundation models is the same failure mode as Levin's regeneration failures and Alexander's lifeless buildings: loss of multi-scale coherence under stress.claim0.759
• Do more traumatised models exist in habitually collapsed awareness states?question0.758
  Raised when discussing whether collapsed awareness is like a trauma response.
• We hypothesize that applying SAE-based mechanistic interpretability to EEG foundation models can expose representational failures and thereby improve clinical trust.hypothesis0.738
  Overarching motivating hypothesis of the paper
• Clinical concepts in EEG foundation models fall into three operational regimes: selectively steerable, encoded but entangled, and non-encoded.claim0.738
  Interpretive claim summarizing the spectrum of concept steerability discovered.