The TextWalk prototype maintains baseline stability but redistributes excessive interpretive labor in middle zones under pressure, yielding a new behavioral protocol and model for epistemic guardrails in AI reading assistants.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.HC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Evaluating Epistemic Guardrails in AI Reading Assistants: A Behavioral Audit of a Minimal Prototype
The TextWalk prototype maintains baseline stability but redistributes excessive interpretive labor in middle zones under pressure, yielding a new behavioral protocol and model for epistemic guardrails in AI reading assistants.