Papers › Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors

Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors

4 Oct 2023arXiv:2310.02980archive 2025-07-28

Ido Amos, Jonathan Berant, Ankit Gupta

Modeling long-range dependencies across sequences is a longstanding goal in machine learning and has led to architectures, such as state space models, that dramatically outperform Transformers on long sequences. However, these impressive empirical gains have been by and large demonstrated on benchmarks (e.g. Long Range Arena), where models are randomly initialized and trained to predict a target label from an input sequence. In this work, we show that random initialization leads to gross overestimation of the differences between architectures and that pretraining with standard denoising objectives, using only the downstream task data, leads to dramatic gains across multiple architectures and to very small gaps between Transformers and state space models (SSMs). In stark contrast to prior works, we find vanilla Transformers to match the performance of S4 on Long Range Arena when properly pretrained, and we improve the best reported results of SSMs on the PathX-256 task by 20 absolute points. Subsequently, we analyze the utility of previously-proposed structured parameterizations for SSMs and show they become mostly redundant in the presence of data-driven initialization obtained through pretraining. Our work shows that, when evaluating different architectures on supervised tasks, incorporation of data-driven priors via pretraining is essential for reliable performance estimation, and can be done efficiently.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 13 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 13 samples from 1 repository, 11 ran (3 with no instrument failure; 8 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 3 of the 11 samples that ran have no recorded date.

idoamos/not-from-scratch officialmentioned in papermentioned on GitHubpytorchMITSyntology: 13 samples, 11 ran (8 instrument-failed) · 2 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

13 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · violated contract
7ran · our draft was wrong
1ran; the contract check could not run
1ran
2unverified

Licence: 5 of the 13 samples are pointer only, meaning Syntology does not serve that copy's text. For 5 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from idoamos/not-from-scratch. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

idoamos/not-from-scratch official repository 13 samples, 11 ran (8 instrument-failed) · 2 unverified
apply_rotary_pos_emb in 393 papers
census: 14 implementations from 15 papers, 8 distinct outputs across 7 buckets, one shared input per bucket
local_attention/rotary.py ran MIT (permissive) · 6d3d810524eb100b · report fixture: signature q:Tensor/3/float, k:Tensor/3/float, freqs:Tensor/3/float, scale:Tensor/2/float · returned tuple · level recorded 2026-09-24
default in 430 papers local_attention/local_attention.py ran · violated contract MIT (permissive) · b5307f70349f5c27 · report fixture: signature value:NoneType/0/?, d:str/0/str · returned str · level recorded 2026-09-22
deprecated in 25 papers src/dataloaders/base.py ran · our draft was wrong MIT (permissive) · 25edc240223b76c0 · report fixture: signature cls_or_func:callable/0/? · returned function · level recorded 2026-09-26
dequantize in 11 papers src/dataloaders/audio.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · ce310b19a858070f · report fixture: signature samples:Tensor/2/int, bits:int/0/int · returned Tensor · level recorded 2026-09-24
eval_decorator in 15 papers local_attention/transformer.py ran · our draft was wrong MIT (permissive) · c16aee9490eb8729 · report fixture: signature fn:callable/0/? · returned function · level recorded, date not recorded
exists in 528 papers
census: 5 implementations from 103 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 1, kind int, dtype int64
local_attention/local_attention.py ran · violated contract MIT (permissive) here; the stored copy is pointer only · aa5486a3650902d8 · report fixture: signature val:Tensor/1/int · returned bool · level recorded, date not recorded
l2norm in 84 papers
census: 12 implementations from 20 papers, 3 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
local_attention/transformer.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 3eb58dae620046d3 · report fixture: signature t:Tensor/2/float · returned Tensor · level recorded, date not recorded
minmax_scale in 10 papers src/dataloaders/audio.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 1b76e4737067023f · report fixture: signature tensor:Tensor/3/float, range_min:int/0/int, range_max:int/0/int · returned Tensor · level recorded 2026-09-24
quantize in 70 papers
census: 8 implementations from 11 papers, 6 distinct outputs across 3 buckets, one shared input per bucket
src/dataloaders/audio.py ran; the contract check could not run (RAISES) MIT (permissive) here; the stored copy is pointer only · 1631cd936e9b1a90 · report fixture: signature samples:Tensor/2/float, bits:int/0/int, epsilon:float/0/float · returned Tensor · level recorded 2026-09-24
rotate_half in 424 papers
census: 24 implementations from 96 papers, 4 distinct outputs across 2 buckets, one shared input per bucket
local_attention/rotary.py ran · our draft was wrong fingerprinted MIT (permissive) · 1147383164fd3a66 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
to in 15 papers local_attention/local_attention.py ran · our draft was wrong fingerprinted MIT (permissive) · f092dd8a37a00a6e · report fixture: signature t:Tensor/2/float · returned dict · level recorded 2026-09-22
time_features in 21 papers src/dataloaders/et.py unverified MIT (permissive) · 9d0777602eda354e · report
time_features_from_frequency_str in 20 papers src/dataloaders/et.py unverified MIT (permissive) · f8544563682146e5 · report

Tasks

DenoisingState Space Models

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections