Is Complex Agentic Memory Always Needed? Naive RAG Is Stronger Than We Thought
The paper tests Raw-RAG, a plain dense-retrieval-plus-reranking baseline over raw conversation history, against agentic memory systems on LoCoMo, LongMemEval, MEME and MemoryAgentBench. Raw-RAG ranks first in 6 of 7 benchmark-model settings and beats the strongest agentic memory system by 0.13 to 5.53 points, which the authors attribute to higher coverage of the evidence needed to derive answers.
Why it matters
A simple baseline outperforming the elaborate agentic-memory stack suggests recent gains on long-horizon conversation memory may be measurement artifacts rather than architectural progress.
Importance: 2/5
notable benchmark result