Systematic experiments reveal that activation steering trades fluency for concept control, is less effective on instruction-tuned models, and that prompting/SFT excel at injection but not removal, with textual metrics correlating to LLM judges.
Unveiling the implicit toxicity in large lan- guage models.arXiv preprint arXiv:2311.17391
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
ZeroUnlearn reformulates machine unlearning as knowledge re-mapping via model editing, using multiplicative updates with closed-form solutions for efficient few-shot removal of sensitive representations while preserving utility.
DKPS-based methods predict new model benchmark scores using cached responses, matching baseline mean absolute error with substantially fewer queries and an offline query selection approach.
AERIC uses a 387-parameter head on LLM hidden states for same-pass anticipatory detection of implicit harm, reporting AUROC gains on DiaSafety and Harmful Advice plus low-latency trigger rates on HarmBench and SocialHarmBench.
citing papers explorer
-
On The Effectiveness-Fluency Trade-Off In LLM Conditioning: A Systematic Study
Systematic experiments reveal that activation steering trades fluency for concept control, is less effective on instruction-tuned models, and that prompting/SFT excel at injection but not removal, with textual metrics correlating to LLM judges.
-
ZeroUnlearn: Few-Shot Knowledge Unlearning in Large Language Models
ZeroUnlearn reformulates machine unlearning as knowledge re-mapping via model editing, using multiplicative updates with closed-form solutions for efficient few-shot removal of sensitive representations while preserving utility.
-
Query-efficient model evaluation using cached responses
DKPS-based methods predict new model benchmark scores using cached responses, matching baseline mean absolute error with substantially fewer queries and an offline query selection approach.
-
AERIC: Anticipatory Hidden-State Monitoring for Implicit Harmful Dialogue
AERIC uses a 387-parameter head on LLM hidden states for same-pass anticipatory detection of implicit harm, reporting AUROC gains on DiaSafety and Harmful Advice plus low-latency trigger rates on HarmBench and SocialHarmBench.