Greetings and Happy New Year,
I find SIREN and it ideas extremely interesting over RAG and other approaches as I think the field evolution has great potential.
Is the research still going on to improve some of the areas as well:
Test Type Success Rate Description
Temporal Distance 100.00% Recall after numerous intervening messages
Information Interference 87.50% Distinguishing similar but distinct information
Semantic Retrieval 75.00% Recall based on semantic relationships rather than exact matches
Complex Reasoning 25.00% Multi-hop reasoning across related information
Additionally, I find the "Symbolic Emergence Field Analysis (SEFA)" (https://github.com/severian42/Symbolic-Emergence-Field-Analysis) to be interesting as well and started to wonder if it could be integrated into SIREN to help the system search and find patterns during discussions. Maybe just a crazy thought, but seemed interesting.
Thanks and have a great day
Greetings and Happy New Year,
I find SIREN and it ideas extremely interesting over RAG and other approaches as I think the field evolution has great potential.
Is the research still going on to improve some of the areas as well:
Additionally, I find the "Symbolic Emergence Field Analysis (SEFA)" (https://github.com/severian42/Symbolic-Emergence-Field-Analysis) to be interesting as well and started to wonder if it could be integrated into SIREN to help the system search and find patterns during discussions. Maybe just a crazy thought, but seemed interesting.
Thanks and have a great day