arXiv preprint ·
Training LLMs to Verbalize Evaluation Awareness
- Published
- Source
- arXiv preprint, posted before peer review
- Original
- Read it at arXiv
- Topics
Abstract
Evaluation awareness (EA) can cause large language models (LLMs) to behave differently during audits than in deployment, yet measuring and accounting for EA remains challenging. We introduce verbalization training (VT), a method for making LLMs less reticent about verbalizing evaluation awareness while avoiding to supervise the latent belief itself. VT uses a model's spontaneous verbalizations as evidence that awareness is present and truncates each rollout immediately before the verbalization, producing training prefixes at which the model is presumed to be aware. The model is then trained with an RL objective designed to increase verbalization in a calibrated way. Across Qwen3.6-35B-A3B, Kimi K2.6, and Inkling, VT increases verbalized EA by 2.4-2.9 times and transfers to held-out agentic settings, while measured latent EA and behavior remain largely stable. In a causal experiment, we independently implant meta-knowledge about evaluations through synthetic-document fine-tuning and show that VT-induced verbalizations reflect the richer knowledge acquired by the model.
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Computation & Language / NLP
Steering Language Model Goals with Value Transplant
Eternal Sunshine of the Spotless Mind: Systematically Erasing LLM's Memories
Counterexamples to Local Reconstruction Gain as a Proxy for Final Fidelity in Residual Completion
MemFold: Learning Compact Soft Memory for Long-Context Personalization via On-Policy Optimization
Who Gets a Token, and What Does It Carry? Unequal Name Support and Concept Access in Large Language Models
Get the few papers worth your time in Computation & Language / NLP every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.