MemRL (Agent / long-term memory): present, but with little supersession signal in the knowledge base. 0 paper(s) critique it, 1 beat it on benchmarks — not ranked as a superseded baseline. Sub-problem: cluster led by Memento. Newer alternatives in the same sub-problem include Learning to Retrieve.

Agent / long-term memory

MemRL

Tracked

Present, but with little supersession signal in the knowledge base

0 papers critique it · 1 beat it on benchmarks

Beaten on benchmarks

Head-to-head results where a newer method reports beating MemRL. Values are copied from the source paper's tables — verify against the cited paper.

Learning to Retrieve beats MemRL
22.17 vs 18.83
Overall · [Phase 2]
Learning to Retrieve: Dual-Level Long-Term Memory for Text-to-SQL Agents

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.

Learning to RetrieveLearning to Retrieve: Dual-Level Long-Term Memory for Text-to-SQL Agents
May 30, 2026