AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Training and Inference on Any-Order Autoregressive Models the Right Way

Andy Shih, Dorsa Sadigh, Stefano Ermon

NeurIPS 2022 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Tractable probabilistic models · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with MAC mask distribution instead of ARDM uniform-cardinality masksARDM · Text8joint bpd: 1.48 -> 1.40Improved
train with MAC instead of ARDMARDM · Text8marginal bpd: 1.12 -> 1.09Improved
train with MAC instead of ARDMARDM · ImageNet32joint bpd: 3.60 -> 3.58Improved
train with MAC instead of ARDMARDM · ImageNet32marginal bpd: 2.10 -> 2.08Improved
train with MAC instead of ARDMARDM · CIFAR10joint bpd: 2.86 -> 2.81Improved
train with MAC instead of ARDMARDM · CIFAR10marginal bpd: 1.84 -> 1.81Improved
add deterministic decomposition protocol only (insight A), no edge weighting (ablation)ARDM · Text8mask distribution ablationImproved
add edge-weighting with random decomposition only (insight B) (ablation)ARDM · Text8mask distribution ablationImproved
combine both insights (w-MAC protocol) (ablation)ARDM · Text8mask distribution ablationImproved
add cardinality reweighing heuristic (ablation)MAC · Text8mask distribution ablationImproved
build MAC on ACE instead of ACE original masksACE · gas conditionallog-likelihood: 9.64 -> 9.77Improved
build MAC on ACE instead of ACE original masksACE · hepmass conditionallog-likelihood: -3.86 -> -3.05Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy