AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Source-Target Inference Models for Spatial Instruction Understanding

Hao Tan, Mohit Bansal

AAAI 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 13 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
add relative-position and stack features to block coordinates (ablation)Full Model (Expectation) · Blank-labeled blocks validationsource accuracy: 50.0% -> 52.2%Improved
add relative-position and stack features to block coordinates (ablation)Full Model (Expectation) · Blank-labeled blocks validationtarget mean distance: 3.41 -> 3.23Improved
replace last-hidden-state attention with CNN attention (ablation)Full Model (Expectation) · Blank-labeled blocks validationsource accuracy: 47.9% -> 52.2%Improved
replace last-hidden-state attention with CNN attention (ablation)Full Model (Expectation) · Blank-labeled blocks validationtarget mean distance: 3.48 -> 3.23Improved
replace CNN attention with dual attention (ablation)Full Model (Expectation) · Blank-labeled blocks validationsource accuracy: 51.4% -> 52.2%No clear effect
replace CNN attention with dual attention (ablation)Full Model (Expectation) · Blank-labeled blocks validationtarget mean distance: 3.32 -> 3.23No clear effect
use expectation inference and loss instead of annealed sampling (ablation)Target position prediction · Blank-labeled blocks validationtarget mean distance: 3.46 -> 3.23Improved
share representations through joint source-target training (ablation)Source block prediction · Blank-labeled blocks validationsource accuracy: 51.3% -> 52.2%Improved
share representations through joint source-target training (ablation)Target position prediction · Blank-labeled blocks validationtarget mean distance: 3.46 -> 3.23Improved
add an 8-sized ensemble to the expectation model (ablation)Full Model (Expectation) · Blank-labeled blocks validationtarget mean distance: 3.23 -> 3.14Improved
combine CNN and dual-attention expectation ensemblesExpectation Model · Blank-labeled blocks testsource accuracy: 56.6% -> 57.70%Improved
initialize the instruction encoder with pretrained GloVe embeddingsInstruction LSTM-RNN · Blank-labeled blocksperformanceNo clear effect

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