A BART-based post-generation corrector lowers toxicity and jailbreaking scores, but the reported gains are partly in-sample because thresholds are optimized on the evaluation data.
Enhancing rag-retrieval to improve llms robustness and resilience to hallucinations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
A Data-Centric Approach for Safe and Secure Large Language Models against Threatening and Toxic Content
A BART-based post-generation corrector lowers toxicity and jailbreaking scores, but the reported gains are partly in-sample because thresholds are optimized on the evaluation data.