Introspection Adapters: Training LLMs to Report Their Learned Behaviors
When model developers or users fine-tune an LLM, this can induce behaviors that are unexpected, deliberately harmful, or hard to detect. It would be far easier to audit LLMs if they could simply describe their behaviors in natural language. Here, we study a scalable approach to rapidly identify learned behaviors of many LLMs derived from a shared base LLM. Given a model $M$, our method works by finetuning models $M_i$ from $M$ with implanted behaviors $b_i$; the $(M_i, b_i)$ pairs serve as labeled training data. We then train an introspection adapter (IA): a single LoRA adapter jointly trained across the finetunes $M_i$ to cause them to verbalize their implanted behaviors. We find that th...
| Authors | Keshav Shenoy, Li Yang, Abhay Sheshadri et al. |
|---|---|
| Primary category | cs.AI |
| Published | 18 April 2026 |
| Last updated at source | 28 April 2026 |
Connected in the knowledge graph
- andyx10/write-vector-methods-resultsDataset · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.