LegközelebbKövetkező útmutató
SWE-pad és kódoló ügynök referenciaértékei
Műszaki
Műszaki ÚTMUTATÓ
Evaluating AI agents means measuring how well a multi-step system reaches goals in realistic settings.
Hallgassa meg ezt az útmutatótAI-narrált · ~3 min
That covers whether it finished the task, whether the path it took was sensible, whether it used tools correctly, what it cost, and whether it stayed safe. Agents act over many steps and have side effects, so checking a single answer against a reference is not enough, and good evaluations usually run the agent inside a controlled environment and inspect the final state.
Agent evaluation looks at several things. Task success asks whether the goal was actually achieved, ideally checked by looking at the environment afterward: tests pass, the right record exists, the file has the right contents. Trajectory quality asks whether the steps were reasonable, without loops, pointless calls or lucky guesses. Tool-use correctness checks whether the agent picked the right tools with valid arguments and handled errors. Cost and latency track tokens, calls and wall-clock time. Safety checks whether the agent avoided harmful or unauthorized actions and resisted prompt injection. Environment-based benchmarks have become a leading approach. SWE-bench, from Princeton researchers, asks agents to resolve real GitHub issues and checks the results with the project's tests, and SWE-bench Verified is a human-checked subset. WebArena provides self-hosted websites for browsing tasks. OSWorld tests agents operating full computer desktops. GAIA poses questions that need multi-step research and tools. The tau-bench suite simulates users and domain policies, and it introduced a pass^k metric that asks whether an agent succeeds on every one of k repeated attempts. Grading uses three main kinds of checks: code-based checks on outcomes, which are precise and cheap; model-based graders that score transcripts against a rubric, which are flexible but need to be checked against human judgment; and human review, which is the most trustworthy and the slowest. A common mistake is to trust public leaderboards alone. Benchmarks can leak into training data, become saturated, or fail to resemble your own workload. Another mistake is treating one run as the truth, since agents are nondeterministic and small differences in success rate can be noise. Teams get the most value from a private evaluation set built from their own real tasks and failures, run repeatedly and tracked over time.
A coding-agent team scores each attempt by whether the repository's hidden tests pass after the agent's patch, as SWE-bench does, instead of judging whether the diff looks right.
A customer-service agent runs against simulated users and a mock booking database. Graders check whether the final database state matches the correct outcome and whether the agent followed refund policy.
An engineering team runs every task five times and reports how often the agent succeeds on all five, which exposes flaky behavior that a single-run success rate would hide.
A company adds a safety suite in which tasks include a planted instruction to delete files or leak data, and it counts an otherwise successful run as failed if the agent obeyed.
As agents take on longer tasks, evaluations are moving toward longer horizons, more realistic environments, and measures of reliability and cost as well as peak capability. Benchmark saturation and contamination will probably keep pushing developers to create new tasks and to rely more on private, domain-specific test sets. Safety and security evaluations, including resistance to prompt injection, are receiving more attention as agents get access to real systems. There is no agreed standard for grading open-ended agent behavior, so combining outcome checks, calibrated model graders and human review is likely to remain common practice.
Ingyenes hírlevél
Minden hétköznap reggel három ellenőrzött AI történet, egyszerű angol nyelven. Örökké ingyen, hirdetések nélkül.
Minden hétköznap egy e-mail. Egy kattintással lemondhat. Soha nem értékesítjük vagy osztjuk meg az Ön címét.
Teszteld magad
Azonnali visszajelzés minden válaszra, és megosztható tanúsítvány egy ellenőrizhető azonosítóval, miután sikeresen teljesít egy kurzust.
Támogassa az ingyenes AI oktatást. AI Understanding egy 501(c)(3) nonprofit – nincs hirdetés, nincs fizetőfal, soha. Tegyen adományt
Evaluating AI agents means measuring how well a multi-step system reaches goals in realistic settings. That covers whether it finished the task, whether the path it took was sensible, whether it used tools correctly, what it cost, and whether it stayed safe. Agents act over many steps and have side effects, so checking a single answer against a reference is not enough, and good evaluations usually run the agent inside a controlled environment and inspect the final state.
Multi-step actions change environments, so you need to check what actually happened and how, not just the final text.
Environment-based checks, here the project's tests, judge whether the fix really works.
Requiring success on all k attempts measures consistency, which matters for agents deployed to real users.
Trajectory checks judge the path taken, catching loops and waste that outcome checks alone miss.
Model graders are flexible but should be checked against human judgment and watched for systematic bias.
Tanulj tovább
További útmutatók készültek ehhez a témához
LegközelebbKövetkező útmutató
SWE-pad és kódoló ügynök referenciaértékei
Műszaki