AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Beyond Text-Only: Towards Multimodal Table Retrieval in Open-World

Da Li, Keping Bi, Jiafeng Guo, Wei Yuan, Fan Yang, Tingting Gao, Xueqi Cheng

ICLR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Multimodal retrieval-augmented modeling

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
represent table content as image instead of OCR text (ablation)Qwen2-VL-2B retriever · TaR-ViRNDCG@5: 65.91 -> 67.42Improved
represent table title as image instead of text (ablation)Qwen2-VL-2B retriever · TaR-ViRRecall@10: 91.69 -> 82.30Hurt
use original web text instead of OCR text (ablation)Qwen2-VL-2B text retriever · TaR-ViRRecall@10 vs NDCG@5: R@10 89.00 -> 90.30; N@5 65.91 -> 65.20Mixed
use image content instead of OCR text on Hard tables (ablation)Qwen2-VL-2B retriever · TaR-ViR Hard tablesNDCG@5: 62.82 -> 65.29Improved
use image content instead of OCR text on Easy tables (ablation)Qwen2-VL-2B retriever · TaR-ViR Easy tablesNDCG@5: 71.86 -> 68.85Hurt
scale VLM2Vec from 2.21B to 7.07B parameters (ablation)VLM2Vec · TaR-ViRNDCG@5: 67.89 -> 70.56Improved
use larger text embedders (Qwen3-Embedding 4.05B) instead of BGE 109M (ablation)text retriever · TaR-ViRNDCG@5: 72.82 -> 71.93Hurt
use multimodal instead of text retriever for RAG with Qwen3-8BRAG pipeline · TaR-ViR QAaccuracy: 58.38 -> 58.55No clear effect
feed table as image instead of text to Qwen2.5-VL-7B generatorRAG pipeline · TaR-ViR QAaccuracy: 38.51 -> 40.87Improved

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