SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.03328 · 2026

StructLens: A Structural Lens for Language Models via Maximum Spanning Trees

Taro Watanabe, Hidetaka Kamigaito, Frederikus Hudi, Yusuke Sakai, Haruki Sakajo, Transformer Block

arXiv · PDF

Code that ran

6 of 7 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 1 no recorded run), measured 2026-11-22; 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/2603.03328.json · badge: /api/ran/2603.03328.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
naist-nlp/structlens canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
convert_tree_format Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
naist-nlp/structlens/src/structlens/metrics.py
pointer only (licence: NONE) · get_code("11473ef9ba8fa9dc")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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.

count_subtree_sizes Ran
ran, honoured the contract we drafted
naist-nlp/structlens/src/structlens/metrics.py
pointer only (licence: NONE) · get_code("b471fd6c8b4bcea5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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.

get_logger Ran
ran; our contract draft was wrong, not the code
naist-nlp/structlens/src/structlens/utils/logging_config.py
pointer only (licence: NONE) · get_code("6d3ecc9fce1c2ffe")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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.

l2_distance Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
naist-nlp/structlens/src/structlens/similarity.py
pointer only (licence: NONE) · get_code("739cb66d97a8b03c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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.

l2_reciprocal_similarity Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
naist-nlp/structlens/src/structlens/similarity.py
pointer only (licence: NONE) · get_code("80659e1fbbca9198")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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.

parent_array_to_tree Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
naist-nlp/structlens/src/structlens/metrics.py
pointer only (licence: NONE) · get_code("99a847c6af296de3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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.

create_masks Not yet run
unverified (harvested, no recorded run)
naist-nlp/structlens/src/structlens/masking.py
pointer only (licence: NONE) · get_code("98938b9ad4dc6e3f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, naist-nlp/structlens 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 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 1 later) and 6 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.

FunctionAlso present inEarliest-datedOthers named
get_logger 8 other papers
4 dated earlier, 1 the same month, 3 later
2111.05897 (2021-11)
Persia: An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters
2401.10491 (2024-01)
Knowledge Fusion of Large Language Models
2412.03187 (2024-12)
Weighted-Reward Preference Optimization for Implicit Model Fusion
and 5 more not named here

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

Language exhibits inherent structures, a property that explains both language acquisition and language change. Given this characteristic, we expect language models to manifest their own internal structures as well. While interpretability research has investigated how models compute representations mechanistically through attention patterns and Sparse AutoEncoders, the organization of the resulting representations is overlooked. To address this gap, we introduce STRUCTLENS, a framework to analyze representations through a holistic structural view. STRUCTLENS constructs maximum spanning trees based on the semantic representations in residual streams, inspired by tree representation in dependency parsing, and provides summaries of token relationships in representation space. We analyze how contiguous tokens are also nearby in representation space and find that middle layers show the strongest local-span organization. Moreover, analysis of pre-training checkpoints reveals that smaller local units become detectable earlier in pre-training, and larger units later. Our findings demonstrate that STRUCTLENS provides insights into how models organize token representations across layers and training. Our code is available at https://github.com/naist-nlp/structlens.

For agents

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

get_harvested_code_for_paper("2603.03328")
get_code_for_paper("2603.03328")
have("2603.03328")

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

curl https://syntology.ai/api/ran/2603.03328.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.03328.svg)](https://syntology.ai/paper/2603.03328)

Connect an agent — have() is free.