Papers › Recurrent Early Exits for Federated Learning with Heterogeneous Clients

Recurrent Early Exits for Federated Learning with Heterogeneous Clients

23 May 2024arXiv:2405.14791archive 2025-07-28

Royson Lee, Javier Fernandez-Marques, Shell Xu Hu, Da Li, Stefanos Laskaridis, Łukasz Dudziak, Timothy Hospedales, Ferenc Huszár, Nicholas D. Lane

Federated learning (FL) has enabled distributed learning of a model across multiple clients in a privacy-preserving manner. One of the main challenges of FL is to accommodate clients with varying hardware capacities; clients have differing compute and memory requirements. To tackle this challenge, recent state-of-the-art approaches leverage the use of early exits. Nonetheless, these approaches fall short of mitigating the challenges of joint learning multiple exit classifiers, often relying on hand-picked heuristic solutions for knowledge distillation among classifiers and/or utilizing additional layers for weaker classifiers. In this work, instead of utilizing multiple classifiers, we propose a recurrent early exit approach named ReeFL that fuses features from different sub-models into a single shared classifier. Specifically, we use a transformer-based early-exit module shared among sub-models to i) better exploit multi-layer feature representations for task-specific prediction and ii) modulate the feature representation of the backbone model for subsequent predictions. We additionally present a per-client self-distillation approach where the best sub-model is automatically selected as the teacher of the other sub-models at each client. Our experiments on standard image and speech classification benchmarks across various emerging federated fine-tuning baselines demonstrate ReeFL's effectiveness over previous works.

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 20 code samples harvested from 1 repository linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 11 unverified. Where Syntology's instrument failed: 3 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: 20 samples from 1 repository, 9 ran (5 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 between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

royson/reefl officialmentioned in paperpytorchSyntology: 20 samples, 9 ran (4 instrument-failed) · 11 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

20 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 11 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.

3ran · our draft was wrong
1ran; the contract check could not run
5ran
11unverified

Licence: 0 of the 20 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 royson/reefl. “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.

royson/reefl official repository 20 samples, 9 ran (4 instrument-failed) · 11 unverified
Adapter_Layer src/models/reefl_vit.py ran · metamorphic tier: invariant MIT (permissive) · 25b4f80e64754f5f · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
Linear in 127 papers src/models/reefl_vit.py ran · metamorphic tier: invariant MIT (permissive) · 00410aa030868e82 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
LoRALinear src/models/reefl_vit.py ran MIT (permissive) · 14af14fc44efac97 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-24
NormAndLinear src/models/reefl_vit.py ran · metamorphic tier: invariant MIT (permissive) · c6f8f65aec8f0cda · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
src/models/reefl_vit.py ran · metamorphic tier: invariant MIT (permissive) · 50c83684a8d032d4 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
get_func_from_config src/models/reefl_vit.py ran · our draft was wrong MIT (permissive) · a0433a63b0aa9b64 · report fixture: signature class_type:dict/0/? · returned type · level recorded 2026-09-22
init_ssf_scale_shift in 9 papers src/models/reefl_vit.py ran · our draft was wrong fingerprinted MIT (permissive) · 9873773b10315cc0 · report fixture: signature dim:int/0/int · returned tuple · level recorded 2026-09-22
prune in 17 papers src/models/reefl_vit.py ran · our draft was wrong MIT (permissive) · 9b6ef57a04624e80 · report fixture: signature state_dict:dict/0/?, param_idx:dict/0/? · returned dict · level recorded 2026-09-22
trunc_normal_ in 154 papers
census: 37 implementations from 54 papers, 4 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
src/models/reefl_vit.py ran; the contract check could not run (RAISES) MIT (permissive) · af8aefa30bd76920 · report fixture: signature tensor:Tensor/2/float, mean:float/0/float, std:float/0/float, a:float/0/float, b:float/0/float · returned Tensor · level recorded 2026-09-22
Mlp in 102 papers src/models/reefl_vit.py unverified MIT (permissive) · 335a33f331dc2ff7 · report
Ree src/models/reefl_vit.py unverified MIT (permissive) · 47dcd63472af63f4 · report
Transformer in 173 papers src/models/reefl_vit.py unverified MIT (permissive) · e16ec4a665b3d4df · report
VisionTransformer in 53 papers src/models/reefl_vit.py unverified MIT (permissive) · dbbfaa54d7e3f160 · report
init_bert_weights src/models/reefl_vit.py unverified MIT (permissive) · 0a5a8b2a981b7267 · report
reefl_vit_template src/models/reefl_vit.py unverified MIT (permissive) · 3abf98fe951b6095 · report
vit_base in 5 papers src/models/reefl_vit.py unverified MIT (permissive) · b50fb2ac893c8ce1 · report
vit_small in 33 papers src/models/reefl_vit.py unverified MIT (permissive) · 1befb46451d50591 · report
vit_tiny in 40 papers src/models/reefl_vit.py unverified MIT (permissive) · c856301a391d8154 · 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 ReeFL (graph method: ReeFL). 27 bricks; individually on a fixture: 16 ran (of which 5 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 5 where Syntology's instrument failed) · 11 unverified. The assembly has not been run end-to-end.

  1. reefl_vit_template entry · level 0 · unverified · 3abf98fe951b6095
  2. get_func_from_config helper · level 1 · ran · our draft was wrong · a0433a63b0aa9b64
  3. prune helper · level 1 · ran · our draft was wrong · 9b6ef57a04624e80
  4. vit_base model · level 0 · unverified · b50fb2ac893c8ce1
  5. vit_small model · level 0 · unverified · 1befb46451d50591
  6. vit_tiny model · level 0 · unverified · c856301a391d8154
  7. VisionTransformer model · level 0 · unverified · dbbfaa54d7e3f160
  8. Block helper · level 0 · unverified · b106d531af0900bc
  9. PatchEmbed helper · level 1 · ran · 50c83684a8d032d4
  10. Ree helper · level 0 · unverified · 47dcd63472af63f4
  11. trunc_normal_ helper · level 1 · ran; the contract check could not run (RAISES) · af8aefa30bd76920
  12. Attention helper · level 0 · unverified · dfbc1c3d50b97f50
  13. DropPath helper · level 1 · ran · ad5f80b54573c0e0
  14. Mlp helper · level 0 · unverified · 335a33f331dc2ff7
  15. drop_path helper · level 1 · ran; the contract check could not run · 55120f2026b56aa2
  16. init_ssf_scale_shift helper · level 1 · ran · our draft was wrong · 9873773b10315cc0
  17. ssf_ada helper · level 1 · ran · honoured contract · 61e8364b7e9bcb70
  18. Linear helper · level 1 · ran · 00410aa030868e82
  19. NormAndLinear helper · level 1 · ran · c6f8f65aec8f0cda
  20. Transformer helper · level 0 · unverified · e16ec4a665b3d4df
  21. Adapter_Layer helper · level 1 · ran · 25b4f80e64754f5f
  22. LoRALinear helper · level 1 · ran · 14af14fc44efac97
  23. FeedForward helper · level 1 · ran · 450f5b303bbc3417
  24. PreNorm helper · level 1 · ran · 9b66b60a7704b891
  25. Residual helper · level 1 · ran · 8907e91277107509
  26. init_bert_weights helper · level 0 · unverified · 0a5a8b2a981b7267
  27. LoRALayer helper · level 1 · ran · 562c6b943203b573

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="2405.14791") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Federated LearningKnowledge DistillationPrivacy Preserving

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Knowledge Distillation

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