mlsanityData to 14 Oct 2026

Shifting the Baseline: Single Modality Performance on Visual Navigation & QA

arXiv 1811.00613 ↗First posted 1 Nov 2018cs.CLcited by 30

Not on the map. Why?

Abstract (arXiv)

We demonstrate the surprising strength of unimodal baselines in multimodal domains, and make concrete recommendations for best practices in future research. Where existing work often compares against random or majority class baselines, we argue that unimodal approaches better capture and reflect dataset biases and therefore provide an important comparison when assessing the performance of multimodal techniques. We present unimodal ablations on three recent datasets in visual navigation and QA, seeing an up to 29% absolute gain in performance over published baselines.

Key facts. Cited by 30 papers in the map (data through 14 Oct 2026). First posted on arXiv 1 Nov 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2018: 1; 2019: 1; 2020: 0; 2021: 6; 2022: 6; 2023: 5; 2024: 6; 2025: 2; 2026: 3.

30 papers.

Sort:

By area:

  1. 2018Reinforced Cross-Modal Matching and Self-Supervised Imitation Learning for Vision-Language Navigationcited by 305
  2. 2021TEACh: Task-driven Embodied Agents that Chatcited by 197
  3. 2024Agent AI: Surveying the Horizons of Multimodal Interactioncited by 170
  4. 2022Vision-and-Language Navigation: A Survey of Tasks, Methods, and Future Directionscited by 117
  5. 2019Help, Anna! Visual Navigation with Natural Multimodal Assistance via Retrospective Curiosity-Encouraging Imitation Learningcited by 100
  6. 2021QA Dataset Explosion: A Taxonomy of NLP Resources for Question Answering and Reading Comprehensioncited by 91
  7. 2022Invariant Grounding for Video Question Answeringcited by 68
  8. 2024ReMEmbR: Building and Reasoning Over Long-Horizon Spatio-Temporal Memory for Robot Navigationcited by 67
  9. 2026MIRAGE: The Illusion of Visual Understandingcited by 65
  10. 2024IsoBench: Benchmarking Multimodal Foundation Models on Isomorphic Representationscited by 38
  11. 2021Diagnosing Vision-and-Language Navigation: What Really Matterscited by 36
  12. 2021Core Challenges in Embodied Vision-Language Planningcited by 30
18 more
  1. 2022Analyzing Generalization of Vision and Language Navigation to Unseen Outdoor Areascited by 24
  2. 2025Enter the Mind Palace: Reasoning and Planning for Long-term Active Embodied Question Answeringcited by 14
  3. 2024Vision-Language Navigation with Embodied Intelligence: A Surveycited by 9
  4. 2023ACQUIRED: A Dataset for Answering Counterfactual Questions In Real-Life Videoscited by 8
  5. 2022Going for GOAL: A Resource for Grounded Football Commentariescited by 7
  6. 2026SpatiaLQA: A Benchmark for Evaluating Spatial Logical Reasoning in Vision-Language Modelscited by 6
  7. 2023Not all Fake News is Written: A Dataset and Analysis of Misleading Video Headlinescited by 5
  8. 2024WorldAPIs: The World Is Worth How Many APIs? A Thought Experimentcited by 4
  9. 2023Which way is `right'?: Uncovering limitations of Vision-and-Language Navigation modelcited by 3
  10. 2021Deep Learning for Embodied Vision Navigation: A Surveycited by 2
  11. 2022Exploring Multi-Modal Representations for Ambiguity Detection & Coreference Resolution in the SIMMC 2.0 Challengecited by 2
  12. 2022Understanding Natural Language in Contextcited by 1
  13. 2024Sequential Compositional Generalization in Multimodal Modelscited by 1
  14. 2026Measuring Language Transfer in Robot Policies: Adding Greek to a Cosmos3 Vision-Language-Action Policycited by 1
  15. 2021Neural Abstructions: Abstractions that Support Construction for Grounded Language Learningcited by 0
  16. 2023Describe me an Aucklet: Generating Grounded Perceptual Category Descriptionscited by 0
  17. 2023Core Challenges in Embodied Vision-Language Planningcited by 0
  18. 2025Can Multitask Learning Enhance Model Explainability?cited by 0