An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.
Python agentic-ragdeep-researchemnlp2024knowledge-curationlarge-language-modelsnaaclnlpreport-generationretrieval-augmented-generation
Update comments.
Yijia Shaob35c27229a52
diff --git a/knowledge_storm/collaborative_storm/modules/co_storm_agents.py b/knowledge_storm/collaborative_storm/modules/co_storm_agents.pyindex afc138d..e092cd6 100644--- a/knowledge_storm/collaborative_storm/modules/co_storm_agents.py+++ b/knowledge_storm/collaborative_storm/modules/co_storm_agents.py@@ -109,8 +109,9 @@ class CoStormExpert(Agent):class SimulatedUser(Agent):"""- Simulated Users is a special type of Agent in Co-STORM that simulate real user interaction behavior or automatic experiments.-+ Simulated Users is a special type of Agent in Co-STORM that simulates real user interaction behavior based on the given intent.++ This class can be used for automatic experiments.For more information, please refer to Section 3.4 of Co-STORM paper: https://www.arxiv.org/pdf/2408.15232"""@@ -318,8 +319,9 @@ class Moderator(Agent):class PureRAGAgent(Agent):"""- PureRAGAgent only handles grounded question generation by retrieving information from retriever based on the query.- It does not utilize any other information besides query itself.+ PureRAGAgent only handles grounded question generation by retrieving information from the retriever based on the query.+ It does not utilize any other information besides the query itself.+It's designed for Co-STORM paper baseline comparison."""