Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-08-17 · source-backed
Principle-Bench releases 168 cryptoasset financial-promotion scenarios mapped to two UK FCA principles with paraphrase, keyword-stuffing and boundary perturbations under a pre-registered rubric (arXiv 2608.14329). The 120B judge that leads on benign inputs collapses from 0.74 to 0.27 accuracy on keyword-stuffed Consumer Duty inputs, and a judge from a different model family agrees with it at only Cohen's kappa 0.16 on that split, localizing the failure to the model rather than the corpus. Across keyword counting, three sentence-transformer embedders, an open-weight judge and a calibrated cascade, no method wins all four axes.
Each link below shares sources, entities, or timing with this story.
Shared entity: Bench / Same source domain / Shared topic / Earlier coverage
Both cover Bench; reported by the same outlet (arxiv.org); overlapping topics (judge, model).
Both cover Bench; reported by the same outlet (arxiv.org); overlapping topics (boundary, model).
Shared entity: Bench / Same source domain / Earlier coverage / Downstream implication
Both cover Bench; reported by the same outlet (arxiv.org); earlier Bench coverage from 2026-08-09.
Shared entity: Bench / Same source domain / Earlier coverage / Tension
Both cover Bench; reported by the same outlet (arxiv.org); earlier Bench coverage from 2026-08-07.
Both cover Bench; reported by the same outlet (arxiv.org); earlier Bench coverage from 2026-08-07.
Both cover Bench; reported by the same outlet (arxiv.org); earlier Bench coverage from 2026-08-06.
Both cover Bench; reported by the same outlet (arxiv.org); earlier Bench coverage from 2026-08-03.
Shared entity: Compliance / Same source domain / Earlier coverage / Tension
Both cover Compliance; reported by the same outlet (arxiv.org); earlier Compliance coverage from 2026-08-02.