Agentic reinforcement learning research
Method and benchmark comparisons
- Credit assignment methods for multi-turn agents. JSON · CSV · BibTeX
- Memory, context management and selective observation in agents. JSON · CSV · BibTeX
- Terminal and software agent benchmark comparison. JSON · CSV · BibTeX
Original passages and task inspection
- Credit assignment in agent reinforcement learning: 3 papers, 24 passages. JSON.
- Terminal benchmark task inspection: 5 papers, 35 passages. JSON.
Search original passages from 8 papers within a 13-paper collection. The remaining 5 entries provide bibliographic information. Each passage links to its original paper and section.