How does the TAE token misalignment threshold impact the factual consistency score of Vicuna-13B versus Baichu

Abstract

Abstract: Abstractive summarization models have demonstrated impressive progress in producing fluent, concise, and human-like summaries. Nevertheless, they also face long-term difficulties, including factual inconsistencies, hallucinations, and misunderstandings of idiomatic expressions, which commonly lead to distortions of intended meanings and incorrect information. To improve the semantic reliability and factual consistency of generated summaries, this paper proposes a new hybrid framework, known as the Non-Idiomatic Abstractive Summarization Model (NIASM), which combines idiomatic substitution and

Research Question

How does the TAE token misalignment threshold impact the factual consistency score of Vicuna-13B versus Baichuan 2 on the FactCC benchmark?

Verification Level

Descriptive public verification status only; aggregate claim counts are public, but individual claim records are not exposed here.

Quality Tier

Descriptive public triage only; this tier does not alter current publication or DOI behavior.

Quality Dimensions

Automated triage signals derived from public fields; not human peer review or independent validation.

Correction Record

Public corrections are additive records. Current status does not claim the synthesis is error-free.

Provenance