AINewsnow

95% recall, 80% answers: what LongMemEval retrieval scores don't tell you

This story is from 2026-10-19. It is preserved in the archive; the latest stories are on the live feed.

Memory layers for AI agents love to quote LongMemEval. You will see 94%, 95%, 96%. Those numbers are real, but they usually measure retrieval: was one of the right sessions in the top 5 results? (recall_any@5). LongMemEval's own metric is different: did the model answer the question correctly? , gr…

Read the full story at DEV Community — AI ↗

Timeline · 1 report

  1. 2026-10-19 02:35 · DEV Community — AI
    95% recall, 80% answers: what LongMemEval retrieval scores don't tell you

More stories

  1. DeepSeek V4.1 Flash 1017 now available on AI Gateway — Vercel Blog
  2. Anthropic commits billions to a 'public benefit fund' — Business Insider AI
  3. Intel unveils its GPU D-series for data centers, featuring up to 480GB of LPDDR5X memory, an unusually large capacity to support demanding inference workloads — Techmeme
  4. Sam Altman's World launches a World ID verification suite for banks, bundling tools to verify human and agent identities, its first industry-specific offering — Techmeme
  5. Anthropic proposes AI safety bill with high-risk thresholds and federal preemption — Politico Technology
  6. AI Gateway logs now show session affinity — Vercel Blog
  7. Consumer apprehension grows as US retailers adopt AI-assisted pricing tools — ABC News Technology
  8. Anthropic gives third-party auditors ’employee-like access’ in AI industry first — ZDNET AI

Get the daily brief of stories like this at 6:30 every morning →