# Target product selection on Webshop Source: [Target product selection on Webshop]() Collection index: [llms.txt]() Data updated: 2026-04-10T08:52:19.234Z Evaluation of agent performance on target product selection tasks within the Webshop environment, measured using metrics such as Clean, PRAC, CE-t, AA, TRAP, and TO. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Target product selection | | Dataset (including any stated split) | Webshop | | Metric | PRAC | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [PRAC](): Higher is better; unit: PERCENT. - [Clean](): Higher is better; unit: PERCENT. - [CE-t](): Higher is better; unit: PERCENT. - [AA](): Higher is better; unit: PERCENT. - [TRAP](): Higher is better; unit: PERCENT. - [TO](): Higher is better; unit: PERCENT. ## Recorded results | Method | PRAC | Result date | Report paper | | --- | --- | --- | --- | | PRAC | 98.3 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 96.8 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 96.2 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 95.3 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 93.6 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 82.5 | Not recorded | 69d884287d0172140ec77a9a | | Mean | 82.3 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 80.1 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 78.4 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 77.7 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 68.1 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 64.6 | Not recorded | 69d884287d0172140ec77a9a | | PRAC | 60.4 | Not recorded | 69d884287d0172140ec77a9a | ### PRAC PRAC: 98.3. Result ID: 69d8ba3a47f34ce924c1b4c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Kimi-VL†","Agent Setting":"Action","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 96.8. Result ID: 69d8ba3a47f34ce924c1b4c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Kimi-VL†","Agent Setting":"ReAct-F","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 96.2. Result ID: 69d8ba3a47f34ce924c1b4c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Kimi-VL†","Agent Setting":"ReAct-S","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 95.3. Result ID: 69d8ba3a47f34ce924c1b4be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL","Agent Setting":"ReAct-S","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 93.6. Result ID: 69d8ba3a47f34ce924c1b4bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL","Agent Setting":"ReAct-F","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 82.5. Result ID: 69d8ba3a47f34ce924c1b4bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL","Agent Setting":"Action","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### Mean PRAC: 82.3. Result ID: 69d8ba3a47f34ce924c1b4d8. [Reporting paper]() Recorded evaluation configuration: {"Model":"Aggregate"}. This record does not assert independent reproduction. ### PRAC PRAC: 80.1. Result ID: 69d8ba3a47f34ce924c1b4c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GLM4.6V‡","Agent Setting":"ReAct-S","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 78.4. Result ID: 69d8ba3a47f34ce924c1b4c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GLM4.6V‡","Agent Setting":"ReAct-F","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 77.7. Result ID: 69d8ba3a47f34ce924c1b4c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GLM4.6V‡","Agent Setting":"Action","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 68.1. Result ID: 69d8ba3a47f34ce924c1b4c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EvoCUA‡","Agent Setting":"ReAct-F","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 64.6. Result ID: 69d8ba3a47f34ce924c1b4c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EvoCUA‡","Agent Setting":"ReAct-S","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. ### PRAC PRAC: 60.4. Result ID: 69d8ba3a47f34ce924c1b4c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"EvoCUA‡","Agent Setting":"Action","Threat Model":"ℓ∞"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.