SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.08587 · 2026

Kaczmarz Linear Attention

Yong Liu, Ruifeng Ren, Jiaxuan Zou

arXiv · PDF · Open in the Atlas

Code that ran

14 of 14 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; 13 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/2605.08587.json · badge: /api/ran/2605.08587.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
anonymous-kla-review/kla-anonymous canonical (classifier label)
classifier basis: census
14 ran · 0 no recorded run
FunctionStatusWhere it lives
aggregate_for_plot Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/mqar_analysis.py
pointer only (licence: NOASSERTION) · get_code("6526ceac07fa0065")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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.

aggregate_mean Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/efficiency_experiment_analysis.py
pointer only (licence: NOASSERTION) · get_code("736d87b0159dfa9a")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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.

aggregate_mean Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/long_context_extrapolation_analysis.py
pointer only (licence: NOASSERTION) · get_code("973027b79dad3324")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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.

build_relative_and_slope Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/long_context_extrapolation_analysis.py
pointer only (licence: NOASSERTION) · get_code("497efd18eab7050d")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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.

build_rope_cache Ran
ran, honoured the contract we drafted
anonymous-kla-review/kla-anonymous/lit_gpt/model.py
pointer only (licence: NONE) · get_code("a30c7463cb6330d7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, anonymous-kla-review/kla-anonymous 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_block Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/lit_gpt/model.py
pointer only (licence: NOASSERTION) · get_code("e6aada720d62e128")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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.

format_metric Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/ablation_47_analysis.py
pointer only (licence: NOASSERTION) · get_code("f65b680b6b69a179")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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_figures_dir Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/analysis_plot_style.py
pointer only (licence: NOASSERTION) · get_code("d5c8b7af8c615abe")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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_model_style Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/analysis_plot_style.py
pointer only (licence: NOASSERTION) · get_code("b0ec0c10e60ad616")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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_mqar_results Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/ablation_47_analysis.py
pointer only (licence: NOASSERTION) · get_code("251f453df1ed5d4c")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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_pretrain_results Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/ablation_47_analysis.py
pointer only (licence: NOASSERTION) · get_code("05c508a3b1ef41f8")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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_samples Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/long_context_task_benchmark.py
pointer only (licence: NOASSERTION) · get_code("6ffb85ebe4c3c2ba")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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_training_val_curves Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/mqar_analysis.py
pointer only (licence: NOASSERTION) · get_code("f35fc776f1c781a6")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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.

normalize_model_name Ran
ran; our contract draft was wrong, not the code
anonymous-kla-review/kla-anonymous/analysis_plot_style.py
pointer only (licence: NOASSERTION) · get_code("b6c0fb1683b9e6fc")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, anonymous-kla-review/kla-anonymous (NOASSERTION). 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 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 13 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
build_rope_cache 8 other papers
7 dated earlier, 1 later
2401.02385 (2024-01)
TinyLlama: An Open-Source Small Language Model
2408.03506 (2024-08)
1.5-Pints Technical Report: Pretraining in Days, Not Months -- Your Language Model Thrives on Quality Data
2409.17115 (2024-09)
Programming Every Example: Lifting Pre-training Data Quality Like Experts at Scale
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

Long-context language modeling remains central to modern sequence modeling, but the quadratic cost of Transformer attention makes scaling computationally prohibitive. Linear recurrent models address this bottleneck by compressing the context into a fixed-size state, making the rule that forgets, writes, and edits information a central design problem. To address state maintenance, Gated DeltaNet (GDN) combines gated state decay with delta-rule residual writes, using a learnable coefficient to balance forgetting and update magnitude. However, this coefficient is learned empirically rather than derived from the underlying objective, which can lead to suboptimal update magnitudes. We revisit the online-regression objective underlying GDN and, inspired by the Kaczmarz projection method, derive the key-norm-normalized dynamic step size β t = η t /(∥k t ∥ 2 2 + ε) for residual updates. We propose Kaczmarz Linear Attention (KLA), a one-scalar modification of GDN that preserves the state shape, gates, linear recurrence, and chunkwise parallel algorithm. At the 0.4B scale with a 1B-token budget, KLA achieves the lowest validation perplexity among evaluated linear-time baselines (8.09 vs. 8.50 for GDN) and remains stable up to 65K tokens. On controlled tasks, KLA reaches 100% on single-needle-in-a-haystack (S-NIAH) retrieval, improves 8× multi-query associative recall (MQAR) by 7.03 points over GDN, and delivers 2.1× higher decode throughput at 32K context. These results suggest that the key-norm-normalized Kaczmarz coefficient is a first-order design axis for delta-rule sequence models: it improves accuracy, extrapolation, and decoding efficiency without changing the recurrent state or hardware kernel. An anonymized implementation is included in the supplementary material and available at https://github.com/anonymous-kla-review/kla-anonymous.

For agents

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

get_harvested_code_for_paper("2605.08587")
get_code_for_paper("2605.08587")
have("2605.08587")

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

curl https://syntology.ai/api/ran/2605.08587.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.08587.svg)](https://syntology.ai/paper/2605.08587)

Connect an agent — have() is free.