SEEDS: Preview-Guided Evidence Admission for Search Agents
To enable such selective control, we introduce SEEDS (Selective Evidence Entry During Search), a preview-guided evidence admission framework for search agents.
ProofPaper ↗
Key points
- Search agents extend large reasoning models with multi-round access to external knowledge, yet most existing systems largely conflate retrieving evidence with consuming it.
- We formulate this overlooked problem as evidence admission, which asks, at each search step, which retrieved candidates should be allowed to enter the agent's evolving reasoning context.
- Across nine knowledge-intensive QA benchmarks under both local corpus and live web search, SEEDS consistently improves search effectiveness while substantially reducing full-evidence consumption.
- These results establish evidence admission as a complementary dimension of agentic search, separating what information is retrieved from what information is ultimately allowed to shape reasoning.
Sources (1)
- [1]SEEDS: Preview-Guided Evidence Admission for Search AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 03:25 AM
To enable such selective control, we introduce SEEDS (Selective Evidence Entry During Search), a preview-guided evidence admission framework for search agents.
Search agents extend large reasoning models with multi-round access to external knowledge, yet most existing systems largely conflate retrieving evidence with consuming it.
Extractive summary: sentences quoted from the sources.