AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

A Closer Look at Generalisation in RAVEN

Steven Spratley, Krista Ehinger, Tim Miller

ECCV 2020 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 31 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 frame-relational 1D convolution encoder to ResNet (Rel-Base)ResNet · RAVEN-10k (all configurations)accuracy (%): 34.51 -> 91.66Improved
add AIR object-centric scene decomposition to Rel-Base (Rel-AIR)Rel-Base · RAVEN-10k (all configurations)accuracy (%): 91.66 -> 94.10Improved
add AIR scene decomposition to Rel-Base, 10% training dataRel-Base · RAVEN (10% of training set)accuracy (%): 24.40 -> 51.39Improved
add AIR scene decomposition to Rel-Base, 25% training dataRel-Base · RAVEN (25% of training set)accuracy (%): 52.24 -> 81.07Improved
add AIR scene decomposition, train Left-Right test Up-DownRel-Base · RAVEN (train Left-Right, test Up-Down)accuracy (%): 32.71 -> 66.77Improved
add AIR scene decomposition, train Up-Down test Left-RightRel-Base · RAVEN (train Up-Down, test Left-Right)accuracy (%): 22.61 -> 60.81Improved
add AIR scene decomposition, train 2x2Grid test 3x3GridRel-Base · RAVEN (train 2x2Grid, test 3x3Grid)accuracy (%): 41.55 -> 67.01Improved
add AIR scene decomposition, train 3x3Grid test 2x2GridRel-Base · RAVEN (train 3x3Grid, test 2x2Grid)accuracy (%): 34.03 -> 61.90Improved
add position and scale latents via bilinear layer instead of linear layer (ablation)Rel-AIR · RAVEN (Left-Right and Up-Down, incl. generalisation between them)accuracy (%): falls to around 43% ± 3 without itImproved
use Rel-Base instead of WReN on PGM extrapolationWReN · PGM extrapolationaccuracy (%): 17.20 -> 22.05Improved
answer-frames-only training exposes bias: accuracy far above chance without context (ablation)answer-only solver · RAVEN (all configurations)accuracy (%): 12.5 -> above 90Improved
shuffle object slots from AIR before the relational encoder (ablation)Rel-AIR · RAVENaccuracy (%)Improved

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