Research map / Vision-language and multimodal / Cross-modal learning
Who Should I Trust? Explicit Confidence-Focused Multimodal Intent Recognition
Yi Liu, Qimeng Yang, Lanlan Lu
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 unimodal confidence weights (vs intermodal only) (ablation) | ECFMIR · MIntRec2.0 | F1 (%): 50.24 -> 53.39 F1 | Improved |
| add cross-modal pairwise confidence weights (vs unimodal only) (ablation) | ECFMIR · MIntRec2.0 | F1 (%): 50.5 -> 53.39 F1 | Improved |
| add focal modulation (ablation) | ECFMIR · MIntRec2.0 | F1 (%): 51.38 -> 53.39 F1 | Improved |
| add geometric calibration (ablation) | ECFMIR · MIntRec2.0 | F1 (%): 52.57 -> 53.39 F1 | Improved |
| add top-k distance loss (ablation) | ECFMIR · MIntRec | ACC (%): 73.03 -> 74.38 ACC | Improved |
| add top-k distance loss (ablation) | ECFMIR · MIntRec2.0 | F1 (%): 50.86 -> 53.39 F1 | Improved |
| use ECFMIR instead of TCL-MAP | ECFMIR · MIntRec | ACC (%): 72.36 -> 74.38 ACC | Improved |
| use ECFMIR instead of strongest baseline | ECFMIR · MIntRec2.0 | ACC (%): 57.75 -> 58.58 ACC | Improved |
| use ECFMIR confidence estimation instead of baselines | ECFMIR · MIntRec calibration | ECE (%): 16.20 (MISA) -> 15.99 ECE | Improved |
Closest papers by meaning
- MIntRec2.0: A Large-scale Benchmark Dataset for Multimodal Intent Recognition and Out-of-scope Detection in Conversations (ICLR 2024 · similarity 0.87)
- Contextual Augmented Global Contrast for Multimodal Intent Recognition (CVPR 2024 · similarity 0.85)
- MultiTrust: A Comprehensive Benchmark Towards Trustworthy Multimodal Large Language Models (NeurIPS 2024 · similarity 0.84)
- Reinforce Trustworthiness in Multimodal Emotional Support System (AAAI 2026 · similarity 0.83)
- Token-Level Contrastive Learning with Modality-Aware Prompting for Multimodal Intent Recognition (AAAI 2024 · similarity 0.83)
- XFeat Revisited: Reproducibility and Evaluation of a Lightweight Image Matcher (NeurIPS 2026 · similarity 0.82)
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.
