Research map / Vision-language and multimodal / Human and embodied modeling
Unsupervised Selection of Negative Examples for Grounded Language Learning
Nisha Pillai, Cynthia Matuszek
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 |
|---|---|---|---|
| select negative examples by Paragraph Vector dissimilarity instead of random sampling | logistic regression visual classifiers · color/shape/object classifiers, 72-object corpus | average classification performance | Improved |
| select negative examples by Paragraph Vector dissimilarity instead of using all other objects as negatives | logistic regression visual classifiers · color/shape/object classifiers, 72-object corpus | average classification performance | Improved |
| random negative sampling versus all other objects as negatives | visual classifiers · 72-object corpus | classification performance | No clear effect |
Closest papers by meaning
- Bridging the Gap: Using Deep Acoustic Representations to Learn Grounded Language from Percepts and Raw Speech (AAAI 2022 · similarity 0.87)
- A Spoken Language Dataset of Descriptions for Speech-Based Grounded Language Learning (NeurIPS 2021 · similarity 0.86)
- Learning Language Structures Through Grounding (AAAI 2025 · similarity 0.86)
- Generating Enhanced Negatives for Training Language-Based Object Detectors (CVPR 2024 · similarity 0.85)
- LaViP: Language-Grounded Visual Prompting (AAAI 2024 · similarity 0.83)
- Large (Vision) Language Models are Unsupervised In-Context Learners (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.
