[2605.00994] Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives
Skip to main content
Learn about arXiv becoming an independent nonprofit.
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
> cs > arXiv:2605.00994
Help | Advanced Search
All fields Title Author Abstract Comments Journal reference ACM classification MSC classification Report number arXiv identifier DOI ORCID arXiv author ID Help pages Full text
Search
GO
quick links
Login
Help Pages
About
-->
Computer Science > Computation and Language
arXiv:2605.00994 (cs)
[Submitted on 1 May 2026]
Title:Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives
Authors:Mohammed Abu Baker, Luca Baroni, Dan Wilhelm
View a PDF of the paper titled Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives, by Mohammed Abu Baker and 2 other authors
View PDF HTML (experimental)
Abstract:Finetuning can significantly modify the behavior of large language models, including introducing harmful or unsafe behaviors. To study these risks, researchers develop model organisms: models finetuned to exhibit specific known behaviors for controlled experimentation. Identifying these behaviors remains challenging. We show that a simple perplexity-based method can surface finetuning objectives from model organisms by leveraging their tendency to overgeneralize their finetuned behaviors beyond the intended context. First, we generate diverse completions from the finetuned model using short random prefills drawn from general corpora. Second, we rank completions by decreasing perplexity gap between reference and finetuned models. The top-ranked completions often reveal the finetuning objectives, without requiring model internals or prior assumptions about the behavior. We evaluate this on a diverse set of model organisms (N=76, 0.5 to 70B parameters), including backdoored models, models finetuned to internalize false facts via synthetic document finetuning, adversarially trained models with hidden concerning behaviors, and models exhibiting emergent misalignment. For the vast majority of model organisms tested, the method surfaces completions revealing finetuning objectives within the top-ranked results, with models trained via synthetic document finetuning or to produce exact phrases being particularly susceptible. We further show that the technique can be effective even without access to the exact pre-finetuning checkpoint: trusted reference models from different families can serve as effective substitutes. As the method requires only next-token probabilities from the finetuned model, it is compatible with API-gated models that expose token logprobs.
Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as: arXiv:2605.00994 [cs.CL]
(or arXiv:2605.00994v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2605.00994
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Luca Baroni [view email] [v1] Fri, 1 May 2026 18:00:55 UTC (1,597 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives, by Mohammed Abu Baker and 2 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new | recent | 2026-05
Change to browse by:
cs cs.AI
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
About
Help
contact arXivClick here to contact arXiv Contact
subscribe to arXiv mailingsClick here to subscribe Subscribe
Copyright
Privacy Policy
Web Accessibility Assistance
arXiv Operational Status