Papers › What Makes Training Multi-Modal Classification Networks Hard?

What Makes Training Multi-Modal Classification Networks Hard?

29 May 2019CVPR 2020 6arXiv:1905.12681archive 2025-07-28

Wei-Yao Wang, Du Tran, Matt Feiszli

Consider end-to-end training of a multi-modal vs. a single-modal network on a task with multiple input modalities: the multi-modal network receives more information, so it should match or outperform its single-modal counterpart. In our experiments, however, we observe the opposite: the best single-modal network always outperforms the multi-modal network. This observation is consistent across different combinations of modalities and on different tasks and benchmarks. This paper identifies two main causes for this performance drop: first, multi-modal networks are often prone to overfitting due to increased capacity. Second, different modalities overfit and generalize at different rates, so training them jointly with a single optimization strategy is sub-optimal. We address these two problems with a technique we call Gradient Blending, which computes an optimal blend of modalities based on their overfitting behavior. We demonstrate that Gradient Blending outperforms widely-used baselines for avoiding overfitting and achieves state-of-the-art accuracy on various tasks including human action recognition, ego-centric action recognition, and acoustic event detection.

PaperPDFConference PDFCodeCode 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 11 code samples harvested from 1 repository linked to this paper: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 11 samples from 1 repository, 5 ran (4 with no instrument failure; 1 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

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.

facebookresearch/R2Plus1D mentioned on GitHubcaffe2Apache-2.0Syntology: 11 samples, 5 ran (1 instrument-failed) · 6 unverified report
facebookresearch/VMZ mentioned on GitHubcaffe2Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Video Classification with Channel-Separated Convolutional Networks (arXiv:1904.02811; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
guide2157/ChulaXrayClassifier mentioned on GitHubSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

11 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 6 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.

1ran · our draft was wrong
4ran
6unverified

Licence: 0 of the 11 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 facebookresearch/R2Plus1D. “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.

facebookresearch/R2Plus1D community (archive-listed) 11 samples, 5 ran (1 instrument-failed) · 6 unverified
accuracy_metric in 7 papers c2/lib/utils/metric.py ran Apache-2.0 (permissive) · eafa75d1a397bede · report fixture: signature softmax:ndarray/1/float, label:int/0/int, k:int/0/int · returned tuple · level recorded 2026-09-25
blobs_from_pkl utilities/model_conversion/convert_models.py ran Apache-2.0 (permissive) · be8f28f90ea3b6fd · report fixture: signature path:PosixPath/0/? · returned dict · level recorded 2026-09-25
create_model in 163 papers c2/lib/models/c3d_model.py ran · our draft was wrong Apache-2.0 (permissive) · 33fb556271040441 · report fixture: signature model:MockModel/0/?, data:str/0/str, num_input_channels:int/0/int, num_labels:int/0/int, label:NoneType/0/?, is_test:int/0/int, no_bias:int/0/int, fc6_dim:int/0/int, fc7_dim:int/0/int · returned str · level recorded 2026-09-22
mean_ap_metric c2/lib/utils/metric.py ran fingerprinted Apache-2.0 (permissive) · 3f1d35c210ec7379 · report fixture: signature predicts:ndarray/2/float, targets:ndarray/2/float · returned tuple · level recorded 2026-09-25
model_validation c2/lib/models/model_builder.py ran Apache-2.0 (permissive) · 4a0058a818794147 · report fixture: signature model_name:str/0/str, model_depth:int/0/int, clip_length:int/0/int, crop_size:int/0/int · returned bool · level recorded 2026-09-25
FlipBGR2RGB c2/lib/utils/model_loader.py unverified Apache-2.0 (permissive) · dfc355d0820eea25 · report
ir_csn_152 utilities/model_conversion/conversion_models.py unverified Apache-2.0 (permissive) · 3c5043c8ffebe8fc · report
r2plus1d_152 pt/vmz/models/r2plus1d.py unverified Apache-2.0 (permissive) · f5d71fa38ad9ad76 · report
r2plus1d_152 utilities/model_conversion/conversion_models.py unverified Apache-2.0 (permissive) · f6417808cd2990b8 · report
r2plus1d_34 pt/vmz/models/r2plus1d.py unverified Apache-2.0 (permissive) · 1793f9b13efb9ae4 · report
r2plus1d_34 utilities/model_conversion/conversion_models.py unverified Apache-2.0 (permissive) · c2d08c4f7298be18 · report

Tasks

Action ClassificationAction RecognitionAction Recognition In VideosClassificationEvent DetectionGeneral ClassificationMulti-modal ClassificationTemporal Action Localization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Classification Kinetics-400 G-Blend (Sports-1M pretrain) Acc@1 78.9 #122 of 207 Archive leaderboard report
Action Classification Kinetics-400 G-Blend Acc@1 77.7 #135 of 207 Archive leaderboard report
Action Recognition In Videos Sports-1M G-Blend Video hit@1 74.8 #1 of 2 Archive leaderboard report
Action Recognition In Videos Sports-1M G-Blend Video hit@5 92.4 #1 of 2 Archive leaderboard report
Action Recognition In Videos miniSports G-Blend Clip Hit@1 49.7 #1 of 1 Archive leaderboard report
Action Recognition In Videos miniSports G-Blend Video hit@1 62.8 #1 of 1 Archive leaderboard report
Action Recognition In Videos miniSports G-Blend Video hit@5 85.5 #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.

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