SOURCE-LINKED INTELLIGENCE

MWE-ECL: Recoverable Long-Range Context Does Not Always Override Local Lexical Priors

arXiv · AI, language, vision and robotics · article · Sep 23, 2026 · UTC

Long-context evaluations often test whether a model can recover distant evidence, but recoverability does not guarantee behavioral influence. We test the prediction that a distant discourse anchor can remain explicitly recoverable yet fail to change the locally preferred reading of a familiar multiword expression; such failures should concentrate when the model's no-anchor default conflicts with the anchor, while prior-correct decisions remain largely preserved. We introduce Multiword Expression Effective Context Length (MWE-ECL), a bilingual diagnostic whose matched anchor-retrieval, no-ancho

Read original source ↗ Open in workspace

recordType

paper

region

Global