Papers › Outlier-Efficient Hopfield Layers for Large Transformer-Based Models

Outlier-Efficient Hopfield Layers for Large Transformer-Based Models

4 Apr 2024arXiv:2404.03828archive 2025-07-28

Jerry Yao-Chieh Hu, Pei-Hsuan Chang, Robin Luo, Hong-Yu Chen, Weijian Li, Wei-Po Wang, Han Liu

We introduce an Outlier-Efficient Modern Hopfield Model (termed OutEffHop) and use it to address the outlier inefficiency problem of {training} gigantic transformer-based models. Our main contribution is a novel associative memory model facilitating outlier-efficient associative memory retrievals. Interestingly, this memory model manifests a model-based interpretation of an outlier-efficient attention mechanism (Softmax₁): it is an approximation of the memory retrieval process of OutEffHop. Methodologically, this allows us to introduce novel outlier-efficient Hopfield layers as powerful alternatives to traditional attention mechanisms, with superior post-quantization performance. Theoretically, the Outlier-Efficient Modern Hopfield Model retains and improves the desirable properties of standard modern Hopfield models, including fixed point convergence and exponential storage capacity. Empirically, we demonstrate the efficacy of the proposed model across large-scale transformer-based and Hopfield-based models (including BERT, OPT, ViT, and STanHop-Net), benchmarking against state-of-the-art methods like Clipped_Softmax and Gated_Attention. Notably, OutEffHop achieves an average reduction of 22+% in average kurtosis and 26+% in the maximum infinity norm of model outputs across four models. Code is available at GitHub (https://github.com/MAGICS-LAB/OutEffHop); models are on Hugging Face Hub (https://huggingface.co/collections/magicslabnu/outeffhop-6610fcede8d2cda23009a98f); future updates are on arXiv (https://arxiv.org/abs/2404.03828).

PaperPDFCodeCode Syntology ranMethod as code

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: 10 ran (of which 5 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 13 samples from 1 repository, 10 ran (5 with no instrument failure; 5 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 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

magics-lab/outeffhop officialmentioned in papermentioned on GitHubpytorchSyntology: 13 samples, 10 ran (5 instrument-failed) · 3 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: 10 ran (of which 5 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

5ran · our draft was wrong
5ran
3unverified

Licence: 0 of the 13 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 magics-lab/outeffhop. “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”. For 3 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

magics-lab/outeffhop official repository 13 samples, 10 ran (5 instrument-failed) · 3 unverified
AlphaChooser STanHop_time_seeries/cross_models/hopfield.py ran MIT (permissive) · e7b18cd260d16db1 · report fixture: returned AlphaChooser · level recorded 2026-09-22
ClipSoftmax STanHop_time_seeries/cross_models/hopfield.py ran fingerprinted MIT (permissive) · 4f7463a0aff83e2a · report fixture: returned ClipSoftmax · level recorded 2026-09-22
EntmaxAlpha STanHop_time_seeries/cross_models/hopfield.py ran fingerprinted MIT (permissive) · 9d1fce5d64cdaaf3 · report fixture: returned EntmaxAlpha · level recorded 2026-09-22
EntmaxBisectFunction STanHop_time_seeries/cross_models/hopfield.py ran MIT (permissive) · d6305b9e14fc507d · report fixture: returned EntmaxBisectFunction · level recorded 2026-09-22
Softmax_1 STanHop_time_seeries/cross_models/hopfield.py ran fingerprinted MIT (permissive) · aba43948313580df · report fixture: returned Softmax_1 · level recorded 2026-09-22
clipped_softmax STanHop_time_seeries/cross_models/hopfield.py ran · our draft was wrong fingerprinted MIT (permissive) · 1e2438ad229df84b · report fixture: signature data:Tensor/2/float, dim:int/0/int, eta:float/0/float, gamma:float/0/float · returned Tensor · level recorded 2026-09-22
clipped_softmax_1 STanHop_time_seeries/cross_models/hopfield.py ran · our draft was wrong fingerprinted MIT (permissive) · e9e2c444e97f90ba · report fixture: signature data:Tensor/2/float, dim:int/0/int, eta:float/0/float, gamma:float/0/float · returned Tensor · level recorded 2026-09-22
entmax_bisect in 6 papers STanHop_time_seeries/cross_models/hopfield.py ran · our draft was wrong MIT (permissive) · 2d3be2464ee960b7 · report fixture: signature X:Tensor/2/float, alpha:float/0/float, dim:int/0/int, n_iter:int/0/int, ensure_sum_one:bool/0/bool · returned Tensor · level recorded 2026-09-22
softmax_1 STanHop_time_seeries/cross_models/hopfield.py ran · our draft was wrong fingerprinted MIT (permissive) · 580329f21ac41e9e · report fixture: signature input:Tensor/2/float, dim:int/0/int · returned Tensor · level recorded 2026-09-22
softmax_n_shifted_zeros STanHop_time_seeries/cross_models/hopfield.py ran · our draft was wrong fingerprinted MIT (permissive) · 9bbad4b677349ef5 · report fixture: signature input:Tensor/2/float, n:int/0/int, dim:int/0/int · returned Tensor · level recorded 2026-09-22
Association STanHop_time_seeries/cross_models/hopfield.py unverified MIT (permissive) · 1b3b7daf7cbbd041 · report
ClipSoftmax_1 STanHop_time_seeries/cross_models/hopfield.py unverified MIT (permissive) · cc3dee79e09eb57e · report
Hopfield STanHop_time_seeries/cross_models/hopfield.py unverified MIT (permissive) · 0e0d879508ab2c99 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for OutEffHop (graph method: OutEffHop). 17 bricks; individually on a fixture: 14 ran (of which 7 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 7 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

  1. Hopfield entry · level 0 · unverified · 0e0d879508ab2c99
  2. Association helper · level 0 · unverified · 1b3b7daf7cbbd041
  3. ClipSoftmax helper · level 1 · ran · 4f7463a0aff83e2a
  4. ClipSoftmax_1 helper · level 0 · unverified · cc3dee79e09eb57e
  5. EntmaxAlpha helper · level 1 · ran · 9d1fce5d64cdaaf3
  6. Softmax_1 helper · level 1 · ran · aba43948313580df
  7. Sparsemax helper · level 1 · ran · 8c413a2b668b9234
  8. clipped_softmax helper · level 1 · ran · our draft was wrong · 1e2438ad229df84b
  9. clipped_softmax_1 helper · level 1 · ran · our draft was wrong · e9e2c444e97f90ba
  10. AlphaChooser helper · level 1 · ran · e7b18cd260d16db1
  11. entmax_bisect helper · level 1 · ran · our draft was wrong · 2d3be2464ee960b7
  12. softmax_1 helper · level 1 · ran · our draft was wrong · 580329f21ac41e9e
  13. SparsemaxFunction helper · level 1 · ran · 31d60d234cc0692f
  14. EntmaxBisectFunction helper · level 1 · ran · d6305b9e14fc507d
  15. softmax_n_shifted_zeros helper · level 1 · ran · our draft was wrong · 9bbad4b677349ef5
  16. flatten_all_but_nth_dim helper · level 1 · ran · our draft was wrong · 32ac3d67805b2251
  17. unflatten_all_but_nth_dim helper · level 1 · ran · our draft was wrong · bdec64ff3f52a1cc

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2404.03828") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

BenchmarkingQuantizationRetrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Benchmarking Wiki-40B OutEffHop-Bert_base Perplexity 6.209 #1 of 1 Archive leaderboard report
Quantization Wiki-40B OutEffHop-Bert_base Perplexity 6.295 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionOPTResidual ConnectionSoftmaxWeight DecayWordPiece

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