Research map / Vision-language and multimodal / Human and embodied modeling
Are Elephants Bigger than Butterflies? Reasoning about Sizes of Objects
Hessam Bagherinezhad, Hannaneh Hajishirzi, Yejin Choi, Ali Farhadi
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 changed | Where | Measured | Effect |
|---|---|---|---|
| replace normal-distribution textual baseline with log-normal size graph model (textual only) | Language-only baseline · Relative size comparisons (486 pairs) | accuracy (%) | Improved |
| add visual observations to textual-only model (ablation) | Our model (textual only) · Relative size comparisons (486 pairs) | accuracy (%) | Improved |
| add textual observations to visual-only model (ablation) | Our model (visual only) · Relative size comparisons (486 pairs) | accuracy (%) | Improved |
| use size graph and likelihood maximization instead of shortest-path products over complete graph | Vision-only baseline · Relative size comparisons (486 pairs) | accuracy (%) | Improved |
| full model with textual and visual vs language-only baseline | Language-only baseline · Relative size comparisons (486 pairs) | accuracy (%): 63.4 -> 83.5 | Improved |
Closest papers by meaning
- Evaluating Multiview Object Consistency in Humans and Image Models (NeurIPS 2024 · similarity 0.84)
- QUANTIPHY: A Quantitative Benchmark Evaluating Physical Reasoning Abilities of Vision-Language Models (CVPR 2026 · similarity 0.83)
- Unveiling the Visual Counting Bottleneck in Vision-Language Models (ICML 2026 · similarity 0.83)
- HumanPCR: Probing MLLM Capabilities in Diverse Human-Centric Scenes (ICLR 2026 · similarity 0.83)
- The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental Design (ICML 2026 · similarity 0.83)
- Mind the GAP: Glimpse-based Active Perception improves generalization and sample efficiency of visual reasoning (ICLR 2025 · similarity 0.83)
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.
