{"slug": "eager-enhancing-generative-event-extraction-via-reinforcement-learning-with", "title": "EAGER: Enhancing Generative Event Extraction via Reinforcement Learning with Verifiable Rewards", "summary": "Researchers introduced EAGER, a reinforcement learning framework for generative event extraction that pairs fine-grained verifiable rewards with Schema-Contrastive Advantage Estimation to counter advantage collapse under sparse binary rewards. EAGER's reward design targets structural validity, extraction accuracy, groundedness, coverage, over-generation, and span precision, and across seven benchmark datasets it outperformed prompting, supervised fine-tuning, and prior reinforcement learning baselines, with a substantial improvement over the strongest prior method. The work is published as arXiv:2609.29230v1.", "body_md": "arXiv:2609.29230v1 Announce Type: new \nAbstract: End-to-end event extraction remains challenging for large language models as it requires simultaneous identification of event triggers, classification of event types, and extraction of schema-grounded argument spans. We present EAGER, a reinforcement learning framework for generative event extraction that combines fine-grained verifiable rewards with Schema-Contrastive Advantage Estimation to alleviate advantage collapse under sparse binary rewards. Our reward design explicitly targets structural validity, extraction accuracy, groundedness, coverage, over-generation, and span precision. Experiments across seven benchmark datasets show that EAGER consistently outperforms prompting, supervised fine-tuning, and prior reinforcement learning baselines, achieving a substantial improvement over the strongest prior method. Results demonstrate that task-aligned verifiable rewards and contrastive advantage estimation substantially improve structured extraction.", "url": "https://wpnews.pro/news/eager-enhancing-generative-event-extraction-via-reinforcement-learning-with", "canonical_source": "https://arxiv.org/abs/2609.29230", "published_at": "2026-09-25 04:00:00+00:00", "updated_at": "2026-09-25 04:01:07.987187+00:00", "lang": "en", "topics": ["natural-language-processing", "large-language-models", "machine-learning", "ai-research"], "entities": ["EAGER", "arXiv", "Schema-Contrastive Advantage Estimation"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/eager-enhancing-generative-event-extraction-via-reinforcement-learning-with", "markdown": "https://wpnews.pro/news/eager-enhancing-generative-event-extraction-via-reinforcement-learning-with.md", "text": "https://wpnews.pro/news/eager-enhancing-generative-event-extraction-via-reinforcement-learning-with.txt", "jsonld": "https://wpnews.pro/news/eager-enhancing-generative-event-extraction-via-reinforcement-learning-with.jsonld"}}