SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.16067 · ICML · 2024

LCA-on-the-Line: Benchmarking Out-of-Distribution Generalization with Class Taxonomies

Francesco Ferroni, Zhiqiu Lin, Shu Kong, Jia Shi, Gautam Gare, Jinjin Tian, Siqi Chai, Arun Vasudevan, Di Feng

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2407.16067.json · badge: /api/ran/2407.16067.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
elvishelvis/lca-on-the-line canonical (classifier label)
classifier basis: classifier_2025
3 ran · 1 no recorded run
FunctionStatusWhere it lives
create_lca_matrix Ran
ran, honoured the contract we drafted
elvishelvis/lca-on-the-line/create_hierarchy.py
pointer only (licence: NONE) · get_code("a553a0729c18f519")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, elvishelvis/lca-on-the-line when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

hierarchical_clustering Ran
ran, honoured the contract we drafted
elvishelvis/lca-on-the-line/create_hierarchy.py
pointer only (licence: NONE) · get_code("ca658df713e1859a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, elvishelvis/lca-on-the-line when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

lca_alignment_loss Ran
ran; our contract draft was wrong, not the code
elvishelvis/lca-on-the-line/linear_probe_loss.py
pointer only (licence: NONE) · get_code("8460a056798aeee7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, elvishelvis/lca-on-the-line when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

construct_latent_hierachy Not yet run
unverified (harvested, no recorded run)
elvishelvis/lca-on-the-line/create_hierarchy.py
pointer only (licence: NONE) · get_code("7c04913ae4face36")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, elvishelvis/lca-on-the-line when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We tackle the challenge of predicting models' Out-of-Distribution (OOD) performance using indistribution (ID) measurements without requiring OOD data. Existing evaluations with "Effective Robustness", which use ID accuracy as an indicator of OOD accuracy, encounter limitations when models are trained with diverse supervision and distributions, such as class labels (Vision Models, VMs, on ImageNet) and textual descriptions (Visual-Language Models, VLMs, on LAION). VLMs often generalize better to OOD data than VMs despite having similar or lower ID performance. To improve the prediction of models' OOD performance from ID measurements, we introduce the Lowest Common Ancestor (LCA)on-the-Line framework. This approach revisits the established concept of LCA distance, which measures the hierarchical distance between labels and predictions within a predefined class hierarchy, such as WordNet. We assess 75 models using ImageNet as the ID dataset and five significantly shifted OOD variants, uncovering a strong linear correlation between ID LCA distance and OOD top-1 accuracy. Our method provides a compelling alternative for understanding why VLMs tend to generalize better. Additionally, we propose a technique to construct a taxonomic hierarchy on any dataset using K-means clustering, demonstrating that LCA distance is robust to the constructed taxonomic hierarchy. Moreover, we demonstrate that aligning model predictions with class taxonomies, through soft labels or prompt engineering, can enhance model generalization. Open source code in our Project Page.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2407.16067")
get_code_for_paper("2407.16067")
have("2407.16067")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2407.16067.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2407.16067.svg)](https://syntology.ai/paper/2407.16067)

Connect an agent — have() is free.