Time-reversed language models, which score queries given responses, improve best-of-N reranking, citation attribution, retrieval, and jailbreak filtering over forward perplexity baselines.
https://github.com/JailbreakBench/jailbreakbench/blob/main/src/jailbreakbench/data/classifier_comparison.csv
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Time-Reversal Provides Unsupervised Feedback to LLMs
Time-reversed language models, which score queries given responses, improve best-of-N reranking, citation attribution, retrieval, and jailbreak filtering over forward perplexity baselines.