REVIEW 2 cited by
Detoxifying Large Language Models via Knowledge Editing
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories with various powerful attack prompts and equips comprehensive metrics for systematic evaluation. We conduct experiments with several knowledge editing approaches, indicating that knowledge editing has the potential to detoxify LLMs with a limited impact on general performance efficiently. Then, we propose a simple yet effective baseline, dubbed Detoxifying with Intraoperative Neural Monitoring (DINM), to diminish the toxicity of LLMs within a few tuning steps via only one instance. We further provide an in-depth analysis of the internal mechanism for various detoxifying approaches, demonstrating that previous methods like SFT and DPO may merely suppress the activations of toxic parameters, while DINM mitigates the toxicity of the toxic parameters to a certain extent, making permanent adjustments. We hope that these insights could shed light on future work of developing detoxifying approaches and the underlying knowledge mechanisms of LLMs. Code and benchmark are available at https://github.com/zjunlp/EasyEdit.
Forward citations
Cited by 2 Pith papers
-
Isotropy Cliffs: The Geometric Signature of Decision-Making in Large Language Models
Across five LLMs, a sharp decrease in embedding isotropy at a critical layer predicts multiple-choice accuracy, with Spearman correlations up to -0.92.
-
Detoxification of Large Language Models through Output-layer Fusion with a Calibration Model
A small calibration model trained on non-toxic text is aligned and fused into the final layer of LLaMA-2-based LLMs, modestly reducing toxicity on RealToxicityPrompts but with mixed perplexity results.
Discussion (0). Sign in to comment.