What it measures
Banking-support conversations combine retrieval from unstructured knowledge with state-changing tool calls.
How to read the result
Look for coordination of knowledge, policy, and the user’s actual request rather than tool syntax alone.
Limits and comparison traps
User simulation, knowledge corpus, and policy versions influence success. Conversation completion does not imply a correct final state.
Benchmark reference notes
Fintech customer-support benchmark from $\tau$-Knowledge testing navigation of large unstructured banking policies and tool execution.
Where this fits in an evaluation suite
Synthetic research rewards do not establish live investment performance. Preserve transaction costs, held-out regimes, constraints, and the exact policy behind a reported result.
Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.
A reproducible comparison
Use the exact implementation, version, task subset, tool access, attempt budget, and judge described by the source.
Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.
PRIMARY EVIDENCE & FURTHER READING
Reviewed 2026-10-26. Explanation and interpretation by RLTasks. Benchmark ownership remains with its original authors.