PairSAE: Mechanistic Interpretability from Pair Representations in Protein Co-Folding
PairSAE summarizes pairwise tensors via N-mode SVD into token-wise interaction roles, then uses a sparse autoencoder to learn shared token-level features that decode into both sequence and pair representations, enabling interpretability of protein co-folding models. Evaluated on Boltz-2 activations, features align with UniProt annotations and predict affinity.
[2606.27440] PairSAE: Mechanistic Interpretability from Pair Representations in Protein Co-Folding
[Submitted on 25 Jun 2026]
Title:PairSAE: Mechanistic Interpretability from Pair Representations in Protein Co-Folding
View a PDF of the paper titled PairSAE: Mechanistic Interpretability from Pair Representations in Protein Co-Folding, by Giosue Migliorini and 5 other authors
View PDF HTML (experimental)
Abstract:Foundation models for structural biology have achieved remarkable performance in predicting biomolecular structure and show promise for the design of proteins and small molecules. Yet understanding which internal features drive their outputs remains challenging. Standard sparse autoencoders (SAEs), effective on transformer-style sequence embeddings, do not transfer cleanly to pairformer-like architectures: naively operating on pairwise representations yields a quadratic blow-up of features and obscures concepts distributed jointly across sequence and pair representations.
We introduce PairSAE, which summarizes pairwise tensors via an N-mode SVD into token-wise interaction roles, then uses a sparse autoencoder to learn a shared set of token-level features that decode into both sequence and pair representations. Evaluated on Boltz-2 activations for PLINDER protein-ligand complexes, PairSAE yields interpretable features that align with UniProt annotations and predict Boltz-2 affinity values. These results indicate that PairSAE links the latent space of foundation models for structural biology to interpretable structural concepts, clarifying what the model "knows" while avoiding pairformer-induced pitfalls that limit conventional SAEs.
Comments: Accepted at the Machine Learning in Structural Biology (MLSB) 2025 workshop
Subjects:
Machine Learning (cs.LG)
Cite as: arXiv:2606.27440 [cs.LG]
(or arXiv:2606.27440v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.27440
arXiv-issued DOI via DataCite
Submission history
From: Giosue Migliorini [view email] [v1] Thu, 25 Jun 2026 18:01:44 UTC (5,248 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled PairSAE: Mechanistic Interpretability from Pair Representations in Protein Co-Folding, by Giosue Migliorini and 5 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.LG
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?)
IArxiv recommender toggle
IArxiv Recommender (What is IArxiv?)
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?)