SARL trains search models to track missing retrieved information
Authors report a 24.3% relative gain over the strongest baseline, with no extra assessment at test time.
Stories per day · 30 days
2 stories, newest first · one card per story
Stories tagged #multi-hop-reasoning
InceptionRAG: Stealthy Poisoning Attack Against Retrieval-Augmented Generation
InceptionRAG fragments malicious payloads into dormant passages that trigger LLMs to self-deduce misinformation via multi-hop reasoning, bypassing existing RAG poisoning defenses.