The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment
The Piggyback Hypothesis proposes that chat-template tokens can piggyback finetuned behavior onto out-of-domain queries. Validated via prefix perturbations, leading to Token-Regularized Finetuning (TReFT) that mitigates emergent misalignment while preserving in-domain learning.
[2606.06667] The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment
[Submitted on 4 Jun 2026]
Title:The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment
View a PDF of the paper titled The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment, by Jiachen Zhao and 5 other authors
View PDF HTML (experimental)
Abstract:The mechanisms behind LLMs' broad over-generalization beyond training examples remain unclear. Emergent misalignment (EM) offers a striking case study: finetuning on narrow tasks induces broad misalignment to semantically-unrelated test domains. In this work, we propose the Piggyback Hypothesis: the chat-template tokens can piggyback the finetuned behaviour onto out-of-domain queries. We validate this hypothesis by showing that subtle perturbations to the prefix (tokens preceding all user queries), or patching the prefix representations with those from the unfinetuned model, can restore alignment without changing the user query. Building on this finding, we propose Token-Regularized Finetuning (TReFT), which regularizes specific token representations during training to mitigate EM. Across different models and multiple EM-inducing datasets, TReFT reduces EM while preserving in-domain learning. On Llama-3.1-8B finetuned on the legal domain, TReFT achieves 33.5% more EM reduction than data interleaving with a retain set of aligned examples. We further show that TReFT extends to other narrow-finetuning settings, including abstention, tool use, and refusal (off-topic generalization is reduced by 54.3% on average), supporting the Piggyback Hypothesis. Broadly, our work highlights that LLMs may learn and generalize in unintended ways and suggests a path toward more constrained finetuning. It also calls for further study of how shared input features can piggyback model behavior across domains.
Subjects:
Computation and Language (cs.CL)
Cite as: arXiv:2606.06667 [cs.CL]
(or arXiv:2606.06667v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.06667
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Jiachen Zhao [view email] [v1] Thu, 4 Jun 2026 19:32:00 UTC (3,472 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment, by Jiachen Zhao and 5 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new | recent | 2026-06
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
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?)