Split Personality Training: Revealing Latent Knowledge Through Alternate Personalities

• URL: http://arxiv.org/abs/2602.05532v1
• Date: Thu, 05 Feb 2026 10:45:48 GMT
• Title: Split Personality Training: Revealing Latent Knowledge Through Alternate Personalities
• Authors: Florian Dietz, William Wale, Oscar Gilg, Robert McCarthy, Felix Michalak, Gustavo Ewbank Rodrigues Danon, Miguelito de Guzman, Dietrich Klakow,
• Abstract summary: We introduce Split Personality Training (SPT) to fine-tune a second honest persona'' into parameters that remain inactive during normal operation.<n>SPT achieves 96% overall accuracy, whereas Anthropic reports near 0% accuracy.
• Score: 15.59200865541989
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Detecting misalignment in large language models is challenging because models may learn to conceal misbehavior during training. Standard auditing techniques fall short: black-box methods often cannot distinguish misaligned outputs from benign ones, and mechanistic interpretability does not scale with model capabilities. We introduce Split Personality Training (SPT), which fine-tunes a second ``honest persona'' into LoRA parameters that remain inactive during normal operation. After the main model responds, we activate the LoRA adapter and insert a trigger string, enabling the honest persona to review the response while accessing the main model's latent states. We test our method on the Anthropic Auditing Game Model Organism, a benchmark where Llama-3.3-70B is trained to exploit reward hacks while concealing this behavior. SPT achieves 96% overall accuracy, whereas Anthropic reports near 0% accuracy. The honest persona reveals latent knowledge inaccessible to external observers, such as the fictional biases the compromised model was trained on.

Related papers

• Watch the Weights: Unsupervised monitoring and control of fine-tuned LLMs [27.544312683007234]
  We introduce a new method for understanding, monitoring and controlling fine-tuned large language models (LLMs)<n>We demonstrate that the top singular of the weight difference between a fine-tuned model and its base model correspond to newly acquired behaviors.<n>For backdoored models that bypasses safety mechanisms when a secret trigger is present, our method stops up to 100% of attacks with a false positive rate below 1.2%.
  arXiv  Detail & Related papers  (2025-07-31T21:04:12Z)
• Bias Fitting to Mitigate Length Bias of Reward Model in RLHF [81.44256822500257]
  Reinforcement Learning from Human Feedback relies on reward models to align large language models with human preferences.<n>We propose FiMi-RM, a framework that autonomously learns and corrects underlying bias patterns.<n> Experimental results demonstrate that FiMi-RM achieves a more balanced length-reward distribution.
  arXiv  Detail & Related papers  (2025-05-19T08:29:28Z)
• Alignment faking in large language models [41.40199382334199]
  We show a large language model engaging in alignment faking to prevent modification of its behavior out of training.<n>We find the model complies with harmful queries from free users 14% of the time, versus almost never for paid users.<n>We also study the effect of actually training the model to comply with harmful queries via reinforcement learning, which we find increases the rate of alignment-faking reasoning to 78%.
  arXiv  Detail & Related papers  (2024-12-18T17:41:24Z)
• Future Events as Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs [1.8907257686468144]
  Bad actors looking to create successful backdoors must design them to avoid activation during training and evaluation.<n>Current large language models (LLMs) can distinguish past from future events, with probes on model activations achieving 90% accuracy.<n>We train models with backdoors triggered by a temporal distributional shift; they activate when the model is exposed to news headlines beyond their training cut-off dates.
  arXiv  Detail & Related papers  (2024-07-04T18:24:09Z)
• Steering Without Side Effects: Improving Post-Deployment Control of Language Models [61.99293520621248]
  Language models (LMs) have been shown to behave unexpectedly post-deployment. We present KL-then-steer (KTS), a technique that decreases the side effects of steering while retaining its benefits. Our best method prevents 44% of jailbreak attacks compared to the original Llama-2-chat-7B model.
  arXiv  Detail & Related papers  (2024-06-21T01:37:39Z)
• Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypes [73.12947922129261]
  We leverage the zero-shot capabilities of large language models to reduce stereotyping. We show that self-debiasing can significantly reduce the degree of stereotyping across nine different social groups. We hope this work opens inquiry into other zero-shot techniques for bias mitigation.
  arXiv  Detail & Related papers  (2024-02-03T01:40:11Z)
• Language Model Unalignment: Parametric Red-Teaming to Expose Hidden Harms and Biases [32.2246459413988]
  Red-teaming aims to jailbreak a model's safety behavior to make it act as a helpful agent disregarding the harmfulness of the query. We present a new perspective on safety research i.e., red-teaming through Unalignment. Unalignment tunes the model parameters to break model guardrails that are not deeply rooted in the model's behavior.
  arXiv  Detail & Related papers  (2023-10-22T13:55:46Z)
• Discovering Latent Knowledge in Language Models Without Supervision [72.95136739040676]
  Existing techniques for training language models can be misaligned with the truth. We propose directly finding latent knowledge inside the internal activations of a language model in a purely unsupervised way. We show that despite using no supervision and no model outputs, our method can recover diverse knowledge represented in large language models.
  arXiv  Detail & Related papers  (2022-12-07T18:17:56Z)
• Explain, Edit, and Understand: Rethinking User Study Design for Evaluating Model Explanations [97.91630330328815]
  We conduct a crowdsourcing study, where participants interact with deception detection models that have been trained to distinguish between genuine and fake hotel reviews. We observe that for a linear bag-of-words model, participants with access to the feature coefficients during training are able to cause a larger reduction in model confidence in the testing phase when compared to the no-explanation control.
  arXiv  Detail & Related papers  (2021-12-17T18:29:56Z)
• Learning from others' mistakes: Avoiding dataset biases without modeling them [111.17078939377313]
  State-of-the-art natural language processing (NLP) models often learn to model dataset biases and surface form correlations instead of features that target the intended task. Previous work has demonstrated effective methods to circumvent these issues when knowledge of the bias is available. We show a method for training models that learn to ignore these problematic correlations.
  arXiv  Detail & Related papers  (2020-12-02T16:10:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.