Papers › Lagging Inference Networks and Posterior Collapse in Variational Autoencoders

Lagging Inference Networks and Posterior Collapse in Variational Autoencoders

16 Jan 2019ICLR 2019 5arXiv:1901.05534archive 2025-07-28

Junxian He, Daniel Spokoyny, Graham Neubig, Taylor Berg-Kirkpatrick

The variational autoencoder (VAE) is a popular combination of deep latent variable model and accompanying variational learning technique. By using a neural inference network to approximate the model's posterior on latent variables, VAEs efficiently parameterize a lower bound on marginal data likelihood that can be optimized directly via gradient methods. In practice, however, VAE training often results in a degenerate local optimum known as "posterior collapse" where the model learns to ignore the latent variable and the approximate posterior mimics the prior. In this paper, we investigate posterior collapse from the perspective of training dynamics. We find that during the initial stages of training the inference network fails to approximate the model's true posterior, which is a moving target. As a result, the model is encouraged to ignore the latent encoding and posterior collapse occurs. Based on this observation, we propose an extremely simple modification to VAE training to reduce inference lag: depending on the model's current mutual information between latent variable and observation, we aggressively optimize the inference network before performing each model update. Despite introducing neither new model components nor significant complexity over basic VAE, our approach is able to avoid the problem of collapse that has plagued a large amount of previous work. Empirically, our approach outperforms strong autoregressive baselines on text and image benchmarks in terms of held-out likelihood, and is competitive with more complex techniques for avoiding collapse while being substantially faster.

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 10 code samples harvested from 1 repository linked to this paper: 7 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 10 samples from 1 repository, 7 ran (6 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.

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.

jxhe/vae-lagging-encoder officialmentioned in papermentioned on GitHubpytorchMITSyntology: 10 samples, 7 ran (1 instrument-failed) · 3 unverified report
spokoyny/spokoyny.github.io 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.

Code Syntology ran Syntology

10 samples harvested: 7 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 1 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.

1ran · our draft was wrong
6ran
3unverified

Licence: 1 of the 10 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, 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 jxhe/vae-lagging-encoder. “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.

jxhe/vae-lagging-encoder official repository 10 samples, 7 ran (1 instrument-failed) · 3 unverified
calc_au image.py ran MIT (permissive) · 6fcc55204af86253 · report fixture: signature model:callable/0/?, test_loader:list/1/tuple, delta:float/0/float · returned tuple · level recorded 2026-09-25
calc_iwnll text.py ran MIT (permissive) · 831c1b0d4145f843 · report fixture: signature model:callable/0/?, test_data_batch:list/1/Tensor, args:Args/0/?, ns:int/0/int · returned tuple · level recorded 2026-09-25
calc_mi image.py ran MIT (permissive) · 9cf81d81989d6812 · report fixture: signature model:callable/0/?, test_loader:list/1/tuple · returned Tensor · level recorded 2026-09-25
calc_mi text.py ran MIT (permissive) · 2d2593fbf606c876 · report fixture: signature model:callable/0/?, test_data_batch:list/1/Tensor · returned Tensor · level recorded 2026-09-25
generate_grid in 14 papers modules/utils.py ran MIT (permissive) · 69f9f574272f9c32 · report fixture: signature zmin:float/0/float, zmax:float/0/float, dz:float/0/float, device:str/0/str, ndim:int/0/int · returned tuple · level recorded 2026-09-25
log_sum_exp in 51 papers
census: 10 implementations from 12 papers, 5 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 2, kind float, dtype float32
modules/utils.py ran · our draft was wrong MIT (permissive) · 5efe6eb2209f6591 · report fixture: signature value:Tensor/2/float, dim:int/0/int, keepdim:bool/0/bool · returned Tensor · level recorded 2026-09-22
test in 377 papers text.py ran MIT (permissive) · 3884f3eb2fb3adc2 · report fixture: signature model:callable/0/?, test_data_batch:list/1/Tensor, mode:str/0/str, args:MockArgs/0/?, verbose:bool/0/bool · returned tuple · level recorded 2026-09-25
get_confirm_token in 41 papers prepare_data.py unverified MIT (permissive) here; the stored copy is pointer only · 135b3dc835ffe6ad · report
test in 377 papers image.py unverified MIT (permissive) · 79e431252f32cf60 · report
test in 377 papers toy.py unverified MIT (permissive) · 3d3feb42f9c79a9e · report

Tasks

Text Generation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Text Generation Yahoo Questions Aggressive VAE KL 5.7 #1 of 3 Archive leaderboard report
Text Generation Yahoo Questions Aggressive VAE NLL 326.7 #1 of 3 Archive leaderboard report
Text Generation Yahoo Questions Aggressive VAE Perplexity 59.7 #1 of 3 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