Research map / Label-efficient and robust learning / Recognition and long-tail
NBCG: Nash-Bargained Causal Game for Long-Tailed Multi-Label NLP
Jing Yang, Jusheng Zhang, Keze Wang
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 |
|---|---|---|---|
| replace Nash bargaining objective with a static weighted sum of coalition losses (ablation) | NBCG · DBpedia Rare-F1 @30% | Rare-label F1: 61.8 -> 65.8 | Improved |
| add Neural SEM causal coalitions instead of random label partitions (ablation) | NBCG · DBpedia Rare-F1 @30% | Rare-label F1: 63.5 -> 65.8 | Improved |
| add adaptive (EMA) disagreement point instead of fixed dk = 0 (ablation) | NBCG · DBpedia Rare-F1 @30% | Rare-label F1: 64.2 -> 65.8 | Improved |
| increase number of players (coalitions) from 1 to 5 (ablation) | NBCG · 20 Newsgroups | mAP: 88.50% -> 92.87% | Improved |
| increase number of players beyond 5 (to 10) (ablation) | NBCG · 20 Newsgroups | mAP: 92.87% -> 90.80% | Hurt |
| use NBCG instead of CCG on high label cardinality samples | CCG · Reuters mAP, cardinality >=6 | mAP: 73.1 -> 78.5 | Improved |
| use NBCG instead of CCG on low cardinality samples | CCG · Reuters mAP, cardinality 1-2 | mAP: 89.2 -> 89.8 | Improved |
| use NBCG instead of RoBERTa fine-tuning for rare labels | RoBERTa · DBpedia Rare-F1 @30% | Rare-label F1: 55.4 -> 65.8 | Improved |
| add NBCG training phases (Neural SEM plus Nash optimization) | RoBERTa (Base) · Reuters training time | relative train time: 1.00x -> 1.15x | Hurt |
| add NBCG to baseline training, head-label performance | RoBERTa · Reuters head labels | label-wise F1 | No clear effect |
Closest papers by meaning
- Sparse Causal Discovery with Generative Intervention for Unsupervised Graph Domain Adaptation (ICML 2025 · similarity 0.83)
- Towards Attributions of Input Variables in a Coalition (ICML 2025 · similarity 0.82)
- From Debate to Equilibrium: Belief‑Driven Multi‑Agent LLM Reasoning via Bayesian Nash Equilibrium (ICML 2025 · similarity 0.82)
- LLM Active Alignment: A Nash Equilibrium Perspective (NeurIPS 2026 · similarity 0.82)
- Learning Coalition Structures with Games (AAAI 2024 · similarity 0.82)
- Uncertainty Herding: One Active Learning Method for All Label Budgets (ICLR 2025 · 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.
