Papers › Classifier-guided Gradient Modulation for Enhanced Multimodal Learning

Classifier-guided Gradient Modulation for Enhanced Multimodal Learning

3 Nov 2024arXiv:2411.01409archive 2025-07-28

Zirun Guo, Tao Jin, Jingyuan Chen, Zhou Zhao

Multimodal learning has developed very fast in recent years. However, during the multimodal training process, the model tends to rely on only one modality based on which it could learn faster, thus leading to inadequate use of other modalities. Existing methods to balance the training process always have some limitations on the loss functions, optimizers and the number of modalities and only consider modulating the magnitude of the gradients while ignoring the directions of the gradients. To solve these problems, in this paper, we present a novel method to balance multimodal learning with Classifier-Guided Gradient Modulation (CGGM), considering both the magnitude and directions of the gradients. We conduct extensive experiments on four multimodal datasets: UPMC-Food 101, CMU-MOSI, IEMOCAP and BraTS 2021, covering classification, regression and segmentation tasks. The results show that CGGM outperforms all the baselines and other state-of-the-art methods consistently, demonstrating its effectiveness and versatility. Our code is available at https://github.com/zrguo/CGGM.

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.

Syntology holds a reference implementation of Classifier-Guided Gradient Modulation (CGGM), generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='Classifier-Guided Gradient Modulation (CGGM)') (how to connect); the free check have('2411.01409') needs no token.

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

Code

Syntology 25 code samples harvested from 1 repository linked to this paper: 17 ran (of which 1 constructed an object rather than computing a result; 9 with no instrument failure: 3 honoured, 1 violated, 5 with no contract checked; 8 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 5 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 25 samples from 1 repository, 17 ran (9 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 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

zrguo/cggm officialmentioned in paperpytorchMITSyntology: 25 samples, 17 ran (8 instrument-failed) · 8 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

25 samples harvested: 17 ran (of which 1 constructed an object rather than computing a result; 9 with no instrument failure: 3 honoured, 1 violated, 5 with no contract checked; 8 where Syntology's instrument failed) · 8 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.

3ran · honoured contract
1ran · violated contract
3ran · our draft was wrong
5ran; the contract check could not run
5ran
8unverified

Licence: 4 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. For 4 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 zrguo/cggm. “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.

zrguo/cggm official repository 25 samples, 17 ran (8 instrument-failed) · 8 unverified
Linear in 127 papers modules/transformer.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 10865bbb99140edd · report fixture: signature in_features:int/0/int, out_features:int/0/int, bias:bool/0/bool · returned Linear · level recorded 2026-09-22
ResNet101 in 11 papers models/segmodel.py ran MIT (permissive) · 6aa213f79f005ec4 · report fixture: signature output_stride:int/0/int, BatchNorm:callable/0/?, input_dim:int/0/int · returned ResNet · level recorded 2026-09-25
augment_gaussian_noise datasets/BratsDataset.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 1fe4d67c3bc4510b · report fixture: signature data_sample:ndarray/2/float, noise_variance:tuple/1/int · returned ndarray · level recorded 2026-09-25
buffered_future_mask in 11 papers modules/transformer.py ran · violated contract fingerprinted MIT (permissive) here; the stored copy is pointer only · 184a5eddedaf6a42 · report fixture: signature tensor:Tensor/2/float, tensor2:Tensor/2/float · returned Tensor · level recorded 2026-09-22
build_aspp in 10 papers models/segmodel.py ran MIT (permissive) · 137ee892a73b5311 · report fixture: signature output_stride:int/0/int, BatchNorm:callable/0/? · returned ASPP · level recorded 2026-09-25
build_decoder in 18 papers models/segmodel.py ran MIT (permissive) · 22ecd809e4d1de39 · report fixture: signature num_classes:int/0/int, BatchNorm:callable/0/?, input_heads:int/0/int · returned Decoder · level recorded 2026-09-25
cal_cos src/foodtrain.py ran · our draft was wrong MIT (permissive) · ed6f4e9e545f8f07 · report fixture: signature cls_grad:list/1/Tensor, fusion_grad:Tensor/1/float · returned list · level recorded 2026-09-24
cutout in 17 papers datasets/BratsDataset.py ran MIT (permissive) · 653ae4081b06cab6 · report fixture: signature image:ndarray/3/int, label:ndarray/2/int, patch_size:tuple/1/int · returned tuple · level recorded 2026-09-25
eval_food src/foodtrain.py ran; the contract check could not run (RAISES) MIT (permissive) · d67ae69618e8f777 · report fixture: signature results:Tensor/2/float, truths:Tensor/1/int · returned float · level recorded 2026-09-22
eval_senti src/eval_metrics.py ran MIT (permissive) · cae47ad9ef47ac82 · report fixture: signature results:Tensor/2/float, truths:Tensor/2/float, exclude_zero:bool/0/bool · returned float · level recorded 2026-09-25
fill_with_neg_inf in 14 papers modules/transformer.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · 8f266b9d616de9b9 · report fixture: signature t:Tensor/2/float · returned Tensor · level recorded 2026-09-22
make_positions in 13 papers modules/position_embedding.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) here; the stored copy is pointer only · 21958d540bd87a8c · report fixture: signature tensor:Tensor/2/int, padding_idx:int/0/int, left_pad:bool/0/bool · returned Tensor · level recorded 2026-09-22
multiclass_acc in 7 papers src/eval_metrics.py ran · honoured contract fingerprinted MIT (permissive) · 65494cdd1508772e · report fixture: signature preds:ndarray/1/float, truths:ndarray/1/float · returned float64 · level recorded 2026-09-25
train_eval_food src/foodtrain.py ran; the contract check could not run (RAISES) MIT (permissive) · b39cbc0e9aa13c48 · report fixture: signature results:Tensor/2/float, truths:Tensor/1/int · returned float · level recorded 2026-09-22
train_eval_iemo src/foodtrain.py ran; the contract check could not run (RAISES) MIT (permissive) · ad9faeb1beaacf40 · report fixture: signature results:Tensor/2/float, truths:Tensor/1/int · returned float · level recorded 2026-09-22
train_eval_senti src/foodtrain.py ran · our draft was wrong MIT (permissive) · 7b0ce64e017cea1a · report fixture: signature results:Tensor/2/float, truths:Tensor/2/float, exclude_zero:bool/0/bool · returned float · level recorded 2026-09-24
weighted_accuracy src/eval_metrics.py ran · honoured contract fingerprinted MIT (permissive) · 6d4da8365bb0f5d1 · report fixture: signature test_preds_emo:ndarray/1/float, test_truth_emo:ndarray/1/float · returned float · level recorded 2026-09-22
Classifier in 72 papers src/foodtrain.py unverified MIT (permissive) · f70f65ad58a53350 · report
ClassifierGuided src/foodtrain.py unverified MIT (permissive) · f927307b92481f41 · report
FoodModel src/foodtrain.py unverified MIT (permissive) · 1cd046b456ad9f31 · report
TransformerEncoder in 96 papers src/foodtrain.py unverified MIT (permissive) · 34ea1f3d71361140 · report
TransformerEncoderLayer in 90 papers src/foodtrain.py unverified MIT (permissive) · 3dfcc20147c77637 · report
initiate src/foodtrain.py unverified MIT (permissive) · e2fe3744d6f3b381 · report
train_model in 103 papers src/foodtrain.py unverified MIT (permissive) · 341db6e7ca441245 · report
train_model in 103 papers src/segtrain.py unverified MIT (permissive) · 27a4ff497fab5848 · 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. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. 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. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

2 assemblies on this paper: 1 harvested, 1 generated.

generated Assembly for Classifier-Guided Gradient Modulation (CGGM) (graph method: Classifier-Guided Gradient Modulation (CGGM)), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 3; 5 property tests passed on both).

label matches by construction: the code was generated for this method (not judged by a model)

  1. compute_modulation_weights entry · level 3 · ran · violated contract · 18c57f195b96b6f7b7f1f993ca1ebdb884d2bb6bbd7e487cd157a0dc10e4ef3a

harvested Assembly for Classifier-Guided Gradient Modulation (CGGM) (graph method: Classifier-Guided Gradient Modulation (CGGM)). 17 bricks; individually on a fixture: 9 ran (of which 3 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. initiate entry · level 0 · unverified · e2fe3744d6f3b381
  2. ClassifierGuided model · level 0 · unverified · f927307b92481f41
  3. FoodModel helper · level 0 · unverified · 1cd046b456ad9f31
  4. train_model helper · level 0 · unverified · 341db6e7ca441245
  5. Classifier helper · level 0 · unverified · f70f65ad58a53350
  6. train_eval_food helper · level 1 · ran; the contract check could not run (RAISES) · b39cbc0e9aa13c48
  7. train_eval_iemo helper · level 1 · ran; the contract check could not run (RAISES) · ad9faeb1beaacf40
  8. train_eval_senti helper · level 1 · ran · our draft was wrong · 7b0ce64e017cea1a
  9. TransformerEncoder helper · level 0 · unverified · 34ea1f3d71361140
  10. cal_cos helper · level 1 · ran · our draft was wrong · ed6f4e9e545f8f07
  11. eval_food helper · level 1 · ran; the contract check could not run (RAISES) · d67ae69618e8f777
  12. LayerNorm helper · level 1 · ran · our draft was wrong · 02408cb7d7b045d1
  13. SinusoidalPositionalEmbedding helper · level 1 · ran · 177caab3f8231bc1
  14. TransformerEncoderLayer helper · level 0 · unverified · 3dfcc20147c77637
  15. make_positions helper · level 1 · ran; the contract check could not run (RAISES) · 21958d540bd87a8c
  16. Linear helper · level 1 · ran · our draft was wrong · 10865bbb99140edd
  17. MultiheadAttention helper · level 0 · unverified · a1bae24ff466241b

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="2411.01409") returns these assemblies with each brick's role and wiring (how to connect).

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