Papers › Towards Scalable and Stable Parallelization of Nonlinear RNNs

Towards Scalable and Stable Parallelization of Nonlinear RNNs

26 Jul 2024arXiv:2407.19115archive 2025-07-28

Xavier Gonzalez, Andrew Warrington, Jimmy T. H. Smith, Scott W. Linderman

Transformers and linear state space models can be evaluated in parallel on modern hardware, but evaluating nonlinear RNNs appears to be an inherently sequential problem. Recently, however, Lim et al. '24 developed an approach called DEER, which evaluates nonlinear RNNs in parallel by posing the states as the solution to a fixed-point problem. They derived a parallel form of Newton's method to solve the fixed-point problem and achieved significant speedups over sequential evaluation. However, the computational complexity of DEER is cubic in the state size, and the algorithm can suffer from numerical instability. We address these limitations with two novel contributions. To reduce the computational complexity, we apply quasi-Newton approximations and show they converge comparably to Newton, use less memory, and are faster. To stabilize DEER, we leverage a connection between the Levenberg-Marquardt algorithm and Kalman smoothing, which we call ELK. This connection allows us to stabilize Newton's method while using efficient parallelized Kalman smoothing algorithms to retain performance. Through several experiments, we show that these innovations allow for parallel evaluation of nonlinear RNNs at larger scales and with greater stability.

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 17 code samples harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 1 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 17 samples from 1 repository, 12 ran (8 with no instrument failure; 4 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 on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 12 samples that ran has no recorded date.

lindermanlab/elk officialmentioned in papermentioned on GitHubjaxBSD-3-ClauseSyntology: 17 samples, 12 ran (4 instrument-failed) · 5 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

17 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 1 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 5 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 · honoured contract
1ran · violated contract
2ran · our draft was wrong
2ran; the contract check could not run
5ran
5unverified

Licence: 0 of the 17 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from lindermanlab/elk. “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.

lindermanlab/elk official repository 17 samples, 12 ran (4 instrument-failed) · 5 unverified
binary_operator in 10 papers elk/models/ssm.py ran; the contract check could not run (RAISES) BSD-3-Clause (permissive) · cc57db8369c5d5df · report fixture: signature q_i:tuple/1/ArrayImpl, q_j:tuple/1/ArrayImpl · returned tuple · level recorded 2026-09-24
discretize_bilinear in 7 papers elk/models/ssm.py ran BSD-3-Clause (permissive) · 9d8d1d6e66ab0b5b · report fixture: signature Lambda:ArrayImpl/1/complex, B_tilde:ArrayImpl/2/complex, Delta:ArrayImpl/1/float · returned tuple · level recorded 2026-09-24
discretize_zoh in 8 papers elk/models/ssm.py ran BSD-3-Clause (permissive) · da7a8a4afbf9a53f · report fixture: signature Lambda:ArrayImpl/1/complex, B_tilde:ArrayImpl/2/complex, Delta:ArrayImpl/1/float · returned tuple · level recorded 2026-09-24
make_DPLR_HiPPO in 5 papers elk/models/ssm_init.py ran · our draft was wrong fingerprinted BSD-3-Clause (permissive) · ccbe2ce2b6220758 · report fixture: signature N:int/0/int · returned tuple · level recorded 2026-09-24
make_HiPPO in 5 papers elk/models/ssm_init.py ran · honoured contract BSD-3-Clause (permissive) · 270b83f06c0fa675 · report fixture: signature N:int/0/int · returned ArrayImpl · level recorded 2026-09-24
make_NPLR_HiPPO in 5 papers elk/models/ssm_init.py ran · our draft was wrong fingerprinted BSD-3-Clause (permissive) · b7812ff1a3838617 · report fixture: signature N:int/0/int · returned tuple · level recorded 2026-09-24
make_scalar_params elk/utils/parallel_kalman_scalar.py ran BSD-3-Clause (permissive) · 5ed5fad0395e9de2 · report fixture: signature initial_mean:Tensor/1/float, dynamics_weights:Tensor/2/float, dynamics_bias:Tensor/2/float, emission_noises:Tensor/2/float · returned ScalarParams · level recorded 2026-09-24
make_unpacking_tuple elk/utils/elk_utils.py ran · honoured contract BSD-3-Clause (permissive) · 139ecb120bda617a · report fixture: signature model:ChainedAF/0/? · returned tuple · level recorded 2026-09-24
rectify in 5 papers elk/models/model_utils.py ran; the contract check could not run (RAISES) BSD-3-Clause (permissive) · 40b10cbf8006f1b1 · report fixture: signature log_scale:ArrayImpl/1/float, softplus:bool/0/bool · returned ArrayImpl · level recorded 2026-09-24
sample_AR_many_layers elk/models/model_utils.py ran BSD-3-Clause (permissive) · 08adf88d4235a6af · report fixture: signature model:SimpleModel/0/?, length:int/0/int, key:ArrayImpl/1/int, dtype:callable/0/? · returned tuple · level recorded 2026-09-24
str2bool in 533 papers elk/models/model_utils.py ran · violated contract BSD-3-Clause (permissive) · 7c508037b40522af · report fixture: signature v:str/0/str · returned bool · level recorded, date not recorded
unpack_state elk/utils/elk_utils.py ran BSD-3-Clause (permissive) · f014c636df2c771c · report fixture: signature unpacking_tuple:list/1/int, packed_state:ArrayImpl/1/complex · returned tuple · level recorded 2026-09-24
deer_iteration elk/algs/deer.py unverified BSD-3-Clause (permissive) · 7b98e2132b6ba6ee · report
deer_iteration_eval elk/algs/deer.py unverified BSD-3-Clause (permissive) · 9c7863de7bd3644d · report
pack_state elk/utils/elk_utils.py unverified BSD-3-Clause (permissive) · 51e5bbc3d32e269a · report
parallel_scalar_filter elk/utils/parallel_kalman_scalar.py unverified BSD-3-Clause (permissive) · a9ec88fdd23680bc · report
seq1d elk/algs/deer.py unverified BSD-3-Clause (permissive) · e76d4d34c46d65cd · report

Tasks

State 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