5 million tokens of user messages.
DolphinBench measures agent memory through tasks, cost, and latency
The benchmark tests whether agents can use long user histories to complete knowledge-work tasks while reporting practical efficiency trade-offs.
Industry
Soumil Rathi · Deshraj Yadav · Taranjeet Singh
Mem0
Research Digest··2 min read
Thread:Memory Management for Agents
Rathi, Yadav, and Singh introduce a benchmark that evaluates long-term memory through agent task completion rather than prompted question answering.
Why this paper
From Mem0 · Part of Memory Management for Agents, now 37 papers
In one line
DolphinBench evaluates agent memory using action-based tasks, cost/latency metrics, and verified tests.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§