Research map / Vision-language and multimodal / Multimodal LLMs
Afford-X: Generalizable and Slim Affordance Reasoning for Task-Oriented Manipulation
Xiaomeng Zhu, Yuyang Li, Leiyao Cui, Pengfei Li, Huan-ang Gao, Yixin Zhu, Hao Zhao
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 |
|---|---|---|---|
| add VA and BF modules to MDETR | MDETR · COCO-Tasks | mAP box: 41.3 -> 43.2 | Improved |
| add VA module (verb-pronoun input) (ablation) | baseline (no VA, BF) · COCO-Tasks verb-pronoun | mAP box: 41.3 -> 43.2 | Improved |
| add BF module (verb-pronoun input) (ablation) | baseline (no VA, BF) · COCO-Tasks verb-pronoun | mAP box: 41.3 -> 43.0 | Improved |
| add both VA and BF (verb-pronoun) (ablation) | baseline · COCO-Tasks verb-pronoun | mAP box: 41.3 -> 43.4 | Improved |
| add both VA and BF (verb-noun) (ablation) | baseline · COCO-Tasks verb-noun | mAP box: 53.2 -> 54.8 | Improved |
| add both VA and BF (distill setting) (ablation) | baseline · COCO-Tasks distill | mAP box: 44.1 -> 45.3 | Improved |
| add noun-pronoun distillation through clustered intermediate features (ablation) | MDETR · COCO-Tasks | mAP box: 41.3 -> 44.1 | Improved |
| distill noun to pronoun directly instead of via cluster features (ablation) | MDETR · COCO-Tasks | mAP box: 44.1 -> 41.9 | Hurt |
| add SBTL alone to baseline (ablation) | MDETR · COCO-Tasks | mAP box: 41.3 -> 43.4 | Improved |
| add CL alone to baseline (ablation) | MDETR · COCO-Tasks | mAP box: 41.3 -> 42.0 | Improved |
| add CCR, CL and SBTL together (ablation) | MDETR · COCO-Tasks | mAP box: 41.3 -> 44.1 | Improved |
| vary cluster number K from 1 to 10 (ablation) | distillation · COCO-Tasks | mAP box | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MDETR — MDETR - Modulated Detection for End-to-End Multi-Modal Understanding (ICCV 2021) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- RAGNet: Large-scale Reasoning-based Affordance Segmentation Benchmark towards General Grasping (ICCV 2025 · similarity 0.93)
- Affordance-R1: Reinforcement Learning for Generalizable Affordance Reasoning in Multimodal Large Language Models (AAAI 2026 · similarity 0.91)
- SeqAfford: Sequential 3D Affordance Reasoning via Multimodal Large Language Model (CVPR 2025 · similarity 0.91)
- AffordDP: Generalizable Diffusion Policy with Transferable Affordance (CVPR 2025 · similarity 0.91)
- AffordBot: 3D Fine-grained Embodied Reasoning via Multimodal Large Language Models (NeurIPS 2025 · similarity 0.90)
- Grounding 3D Object Affordance with Language Instructions, Visual Observations and Interactions (CVPR 2025 · similarity 0.89)
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.
