Do Small Language Models Know When They're Wrong? Confidence-Based Cascade Scoring for Educational Assessment

arXiv 2604.19781•68fdd2da694fade1845563eb5d8f7e37d30af1c08522590073b0403d57666345
AI companionsLLM misusealignmentbehavioral addictionbiological weaponizationbiosecuritycascade systemsconfidence calibrationeducation governancefairnesshiring biasincident monitoringmodel governancemodel safetymulti-agent systemspublic-health surveillancesoft-label governance

Paper metadata

arXiv ID
2604.19781
Version
Not specified by this published record
Category
Computer Science — Computers and Society (cs.CY)

The PDF link points to arxiv.org. Baitaphish does not expose a private stored PDF.

Evidence and limitations

Source ID
arxiv_cs_cy
Record identifier
68fdd2da694fade1845563eb5d8f7e37d30af1c08522590073b0403d57666345
Enrichment time
2026-04-23T07:23:54Z
AI-assisted enrichment
Yes

This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.