Research map / Label-efficient and robust learning / Domain adaptation and generalization
De-biased Natural Language Egocentric Task Verification via Prototypical Evidence Learning
Chong Liu, Xun Jiang, Fumin Shen, Lei Zhu, Jingkuan Song, Heng Tao Shen, Xing Xu
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 PEL | NSG · EgoCross | Macro-F1: 51.7 -> 61.9 | Improved |
| add PEL | PHGC · EgoCross | Macro-F1: 53.4 -> 66.2 | Improved |
| add MEC module | NSG · EgoTV | Macro-F1: 73.9 -> 76.8 | Improved |
| add MEC module | PHGC · EgoTV | Macro-F1: 81.2 -> 83.8 | Improved |
| add Cross-Domain Alignment to IDP+MEC (ablation) | PEL · EgoCross | Macro-F1: 62.0 -> 66.2 | Improved |
| add Intra-Domain Preservation to CDA+MEC (ablation) | PEL · EgoCross | Macro-F1: 62.8 -> 66.2 | Improved |
| add MEC to CDA+IDP (ablation) | PEL · EgoCross | Macro-F1: 64.0 -> 66.2 | Improved |
| add MEC alone to baseline (ablation) | PHGC · EgoCross | Macro-F1: 52.6 -> 54.9 | Improved |
| add CDA alone to baseline (ablation) | PHGC · EgoCross | Macro-F1: 52.6 -> 61.3 | Improved |
| increase anchors K from 30 to 40 (ablation) | PEL · EgoCross Lab | Macro-F1: 48.0 -> 54.1 | Improved |
| increase anchors K from 40 to 50 (ablation) | PEL · EgoCross Lab | Macro-F1: 54.1 -> 49.4 | Hurt |
| use anchors (K=40) vs none (K=0) (ablation) | PEL · EgoCross Kitchen | Macro-F1: 53.0 -> 68.3 | 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.
- Otter — OTTER: Effortless Label Distribution Adaptation of Zero-shot Models (NeurIPS 2024) · better on 2 of 2 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
- PHGC: Procedural Heterogeneous Graph Completion for Natural Language Task Verification in Egocentric Videos (CVPR 2025 · similarity 0.87)
- X-MIC: Cross-Modal Instance Conditioning for Egocentric Action Generalization (CVPR 2024 · similarity 0.84)
- EgoThinker: Unveiling Egocentric Reasoning with Spatio-Temporal CoT (NeurIPS 2025 · similarity 0.84)
- Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention (NeurIPS 2025 · similarity 0.84)
- Test-Time Adaptation for Combating Missing Modalities in Egocentric Videos (ICLR 2025 · similarity 0.84)
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs (NeurIPS 2025 · similarity 0.84)
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.
