# Dexterous Manipulation on D4RL Adroit human cloned v2 Source: [Dexterous Manipulation on D4RL Adroit human cloned v2]() Collection index: [llms.txt]() Data updated: 2026-05-08T10:07:56.068Z Evaluation of dexterous manipulation policies on the D4RL Adroit human cloned v2 dataset, reporting performance metrics (P, H, D) against human and cloned expert baselines, and the overall average. 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 | Dexterous Manipulation | | Dataset (including any stated split) | D4RL Adroit human cloned v2 | | Metric | Performance (Human Expert) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Performance (Human Expert)](): Higher is better; unit: PERCENT. - [Human Likeness (Human Expert)](): Higher is better; unit: PERCENT. - [Distance (Human Expert)](): Lower is better; unit: SCALAR. - [Performance (Cloned Expert)](): Higher is better; unit: PERCENT. - [Human Likeness (Cloned Expert)](): Higher is better; unit: PERCENT. - [Distance (Cloned Expert)](): Lower is better; unit: SCALAR. - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Performance (Human Expert) | Result date | Report paper | | --- | --- | --- | --- | | SlimDT | 94.4 | Not recorded | 69fd6e155497b3b670c89419 | | DC | 93.8 | Not recorded | 69fd6e155497b3b670c89419 | | GDT | 92.5 | Not recorded | 69fd6e155497b3b670c89419 | | DHi | 86.6 | Not recorded | 69fd6e155497b3b670c89419 | | DT | 79.5 | Not recorded | 69fd6e155497b3b670c89419 | | StAR | 77.9 | Not recorded | 69fd6e155497b3b670c89419 | | IQL | 71.5 | Not recorded | 69fd6e155497b3b670c89419 | | BCQ | 66.9 | Not recorded | 69fd6e155497b3b670c89419 | | BC | 63.9 | Not recorded | 69fd6e155497b3b670c89419 | | CQL | 37.5 | Not recorded | 69fd6e155497b3b670c89419 | ### SlimDT Performance (Human Expert): 94.4. Result ID: 69fdb46c47f34ce924e127ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"condition injector":"cross-attention","variant":"pre-cond"}. This record does not assert independent reproduction. ### DC Performance (Human Expert): 93.8. Result ID: 69fdb46c47f34ce924e127c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GDT Performance (Human Expert): 92.5. Result ID: 69fdb46c47f34ce924e127c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DHi Performance (Human Expert): 86.6. Result ID: 69fdb46c47f34ce924e127c9. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DT Performance (Human Expert): 79.5. Result ID: 69fdb46c47f34ce924e127c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### StAR Performance (Human Expert): 77.9. Result ID: 69fdb46c47f34ce924e127c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### IQL Performance (Human Expert): 71.5. Result ID: 69fdb46c47f34ce924e127c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BCQ Performance (Human Expert): 66.9. Result ID: 69fdb46c47f34ce924e127c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BC Performance (Human Expert): 63.9. Result ID: 69fdb46c47f34ce924e127c4. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CQL Performance (Human Expert): 37.5. Result ID: 69fdb46c47f34ce924e127c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.