SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.19944 · 2026

A Measure-Theoretic Analysis of Reasoning: Structural Generalization and Approximation Limits

Yuyang Zhang, Xiaoyin Chen, Xuehai Zhou, Yifu Zhang

arXiv · PDF · Open in the Atlas

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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 3 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/2605.19944.json · badge: /api/ran/2605.19944.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
Yyuzrah/AMTA4R canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
combine_nums Ran
ran; our contract draft was wrong, not the code
Yyuzrah/AMTA4R/MTA4R/src/countdown_utils.py
pointer only (licence: NONE) · get_code("088b5a4931c549e9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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.

compute_metrics Ran
ran; our contract draft was wrong, not the code
Yyuzrah/AMTA4R/MTA4R/src/distribution_metrics.py
pointer only (licence: NONE) · get_code("029738e0501d1cd4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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.

extract_structural_features Ran
ran, honoured the contract we drafted
Yyuzrah/AMTA4R/MTA4R/src/distribution_metrics.py
pointer only (licence: NONE) · get_code("609b16107d5cece9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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.

load_and_featurize Ran
ran, honoured the contract we drafted
Yyuzrah/AMTA4R/MTA4R/src/distribution_metrics.py
pointer only (licence: NONE) · get_code("7ba5b63ac57d388d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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.

mult_heuristic Ran
ran; our contract draft was wrong, not the code
Yyuzrah/AMTA4R/MTA4R/src/countdown_utils.py
pointer only (licence: NONE) · get_code("3b9eb49d3b384506")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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.

sum_heuristic Ran
ran, violated the contract we drafted
Yyuzrah/AMTA4R/MTA4R/src/countdown_utils.py
pointer only (licence: NONE) · get_code("221eded453668b01")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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.

load_model_and_tokenizer Not yet run
unverified (harvested, no recorded run)
Yyuzrah/AMTA4R/MTA4R/src/eval_flexible.py
pointer only (licence: NONE) · get_code("865ff8f8286bfd35")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Yyuzrah/AMTA4R 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, 3 are also present in other papers whose code we hold (3 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.

FunctionAlso present inEarliest-datedOthers named
combine_nums 3 other papers
3 dated earlier
2404.03683 (2024-04)
Stream of Search (SoS): Learning to Search in Language
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.15466 (2025-04)
Learning Adaptive Parallel Reasoning with Language Models
sum_heuristic 3 other papers
3 dated earlier
2404.03683 (2024-04)
Stream of Search (SoS): Learning to Search in Language
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.15466 (2025-04)
Learning Adaptive Parallel Reasoning with Language Models
mult_heuristic 2 other papers
2 dated earlier
2404.03683 (2024-04)
Stream of Search (SoS): Learning to Search in Language
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training

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

While empirical scaling laws for LLM reasoning are well-documented, the theoretical mechanisms governing out-of-distribution (OOD) generalization remain elusive. We formalize reasoning via optimal transport, projecting discrete trajectories into a continuous metric space to quantify domain shifts using the Wasserstein-1 distance. Invoking Kantorovich duality, we bound OOD generalization via architectural Lipschitz continuity and functional approximation limits. This exposes two primary constraints. First, position-dependent attention (e.g., Absolute Positional Encoding) fails to preserve shift invariance, yielding an Ω(1) Lipschitz constant and expected risk, whereas shift-invariant mechanisms (e.g., Rotary Embeddings) preserve equivariance and bound the error. Second, by mapping sequential backtracking to a Dyck-k language, we establish a strict circuit depth lower bound for TC 0 Transformers. Scaling physical layer depth is necessary to avert representation collapse-a constraint that scaling representation width cannot bypass due to irreducible approximation bounds in Barron spaces. Evaluations across 54 Transformer configurations on combinatorial search corroborate these bounds, demonstrating that generalization risk degrades monotonically with the Wasserstein domain shift.

For agents

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

get_harvested_code_for_paper("2605.19944")
get_code_for_paper("2605.19944")
have("2605.19944")

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

curl https://syntology.ai/api/ran/2605.19944.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.19944.svg)](https://syntology.ai/paper/2605.19944)

Connect an agent — have() is free.