Work

Honest Abstention Transfer in Small-Corpus Supervised Fine-Tuning: Measured Refusals and Confabulations in a 4B Estate-Tuned Model

Year: 2026

Type: preprint

Abstract: Measures abstention transfer in a production 4B estate-tuned model with an 8-probe held-out suite. Reports verbatim: correct refusals for unmeasured-value probes AND confabulated citations on citation-shaped probes. Defines the eval-driven DPO counterweight protocol built from measured failures.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.24005478

Author: Lutar, Stephen P., Jr.

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7225413020

Cites:

Cited by:

Related to: 10