SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2512.11784 · ICML · 2026

Softmax as Linear Attention in the Large-Prompt Regime: a Measure-based Perspective

Etienne Boursier, Claire Boyer

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 6 failures of our instrument; 0 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/2512.11784.json · badge: /api/ran/2512.11784.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
eboursier/softmax_as_linear canonical (classifier label)
classifier basis: census
9 ran · 0 no recorded run
FunctionStatusWhere it lives
data_generator Ran
ran; our contract draft was wrong, not the code
eboursier/softmax_as_linear/utils/funcs.py
pointer only (licence: NONE) · get_code("77e156a5d0ccd851")
How it ran

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

find_latest_checkpoint Ran
ran; our contract draft was wrong, not the code
eboursier/softmax_as_linear/utils/data_utils.py
pointer only (licence: NONE) · get_code("d5a7e5d3fda38361")
How it ran

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

generate_covariance_matrix Ran
ran; our contract draft was wrong, not the code
eboursier/softmax_as_linear/utils/data_utils.py
pointer only (licence: NONE) · get_code("b6785078c117c134")
How it ran

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

parse_kcm Ran
ran, honoured the contract we drafted
eboursier/softmax_as_linear/utils/data_utils.py
pointer only (licence: NONE) · get_code("6a9784cbc1331433")
How it ran

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

ridge_estimator Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
eboursier/softmax_as_linear/utils/funcs.py
pointer only (licence: NONE) · get_code("da9431f5172c6d7a")
How it ran

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

ridge_loss Ran
ran, honoured the contract we drafted
eboursier/softmax_as_linear/utils/funcs.py
pointer only (licence: NONE) · get_code("6d1c024dcfc3e22d")
How it ran

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

train_step Ran
ran; our contract draft was wrong, not the code
eboursier/softmax_as_linear/run_experiments.py
pointer only (licence: NONE) · get_code("e770dd61f4f4c86a")
How it ran

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

validate_step Ran
ran; our contract draft was wrong, not the code
eboursier/softmax_as_linear/run_experiments.py
pointer only (licence: NONE) · get_code("1658d189404d61fa")
How it ran

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

weight_l2_distance Ran
ran, honoured the contract we drafted
eboursier/softmax_as_linear/utils/plot_utils.py
pointer only (licence: NONE) · get_code("c2c2717b1ad154a9")
How it ran

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

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 9 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

Softmax attention is a central component of transformer architectures, yet its nonlinear structure poses significant challenges for theoretical analysis. We develop a unified, measure-based framework for studying single-layer softmax attention under both finite and infinite prompts. For i.i.d. Gaussian inputs, we lean on the fact that the softmax operator converges in the infiniteprompt limit to a linear operator acting on the underlying input-token measure. Building on this insight, we establish non-asymptotic concentration bounds for the output and gradient of softmax attention, quantifying how rapidly the finite-prompt model approaches its infiniteprompt counterpart, and prove that this concentration remains stable along the entire training trajectory in general in-context learning settings with sub-Gaussian tokens. In the case of incontext linear regression, we use the tractable infinite-prompt dynamics to analyze training at finite prompt length. Our results allow optimization analyses developed for linear attention to transfer directly to softmax attention when prompts are sufficiently long, showing that largeprompt softmax attention inherits the analytical structure of its linear counterpart. This, in turn, provides a principled and broadly applicable toolkit for studying the training dynamics and statistical behavior of softmax attention layers in large prompt regimes.

For agents

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

get_harvested_code_for_paper("2512.11784")
get_code_for_paper("2512.11784")
have("2512.11784")

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

curl https://syntology.ai/api/ran/2512.11784.json

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

[![Syntology run record](https://syntology.ai/api/ran/2512.11784.svg)](https://syntology.ai/paper/2512.11784)

Connect an agent — have() is free.