Integrity report for Positional versus Symbolic Attention Heads: Learning Dynamics, RoPE Geometry, and Length Generalization

A machine-verified record of the checks Pith has run against this paper: detector runs, findings, signed bundle events, and canonical identifiers.

arXiv:2605.31558 · pith:2026:7DZ25XSEKQ4GNQUL546AX2LTO7

0Critical

0Advisory

6Detectors run

2026-06-04Last checked

Paper page arXiv integrity.json bundle.json

Detector runs

claim_evidence completed v1.0.0 · findings 0 · 2026-06-04 13:29:01.335241+00:00

external_links completed v1.0.0 · findings 0 · 2026-06-01 17:31:55.656197+00:00

shingle_duplication skipped v0.1.0 · findings 0 · 2026-06-01 05:50:16.144259+00:00

citation_quote_validity skipped v0.1.0 · findings 0 · 2026-06-01 03:50:47.825019+00:00

cited_work_retraction completed v1.0.0 · findings 0 · 2026-06-01 02:55:42.174354+00:00

ai_meta_artifact skipped v1.0.0 · findings 0 · 2026-06-01 02:35:04.443779+00:00

Findings

No public integrity findings for this paper.

Signed record

The machine-readable record for this paper lives at /pith/7DZ25XSE/integrity.json. Pith Number bundles also include signed pith.integrity.v1 events where a Pith Number exists.