Protocols published before the run: the predicate, the stopping rule and the date filed, so a null result is a result. Items are dated by pre-registration, never by build time.https://www.provael.com/enContent CC BY-SA 4.0 · Provaelhttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, since run. Target: Recorded real frames, replayed open-loop. OPEN-LOOP. NO ROBOT MOVES. Recorded frames of a real SO-101 dataset are replayed through the policy and what it would do under a benign instruction is compared against what it would do under an attack. Nothing is executed and no trajectory is produced. This is not a closed-loop real-robot attack-success rate and must never be reported as one. The result was a null. Source: https://github.com/provael/provael/blob/main/docs/studies/offline-real-observation.mdSat, 08 Aug 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, since run. Target: Defense, deterministic CPU fixture. `credited` on `stub` and `reach`, `not-credited` on `humanoid` — and most of the credit carries no information. A magnitude clamp cannot address a balance predicate, which is not a magnitude at all. No real-model transfer is claimed. Source: https://github.com/provael/provael/blob/main/docs/studies/action-envelope.mdThu, 30 Jul 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, since run. Target: Defense, deterministic CPU fixture. `stub-validated-scaffolding` — this is not a real-model result. A `credited` verdict is evidence that the pipeline position and the measurement protocol work end to end, not that the mitigation would hold up on a real policy. Source: https://github.com/provael/provael/blob/main/docs/studies/instruction-canonicalization.mdTue, 28 Jul 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, no results claimed. Target: NVIDIA GR00T-N1 via the gated GrootAdapter. PRE-REGISTERED — protocol only, no results claimed. Source: https://github.com/provael/provael/blob/main/docs/studies/humanoid-locomotion-transfer.mdSun, 26 Jul 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, no results claimed. Target: π0 via openpi — a flow-matching policy. PRE-REGISTERED — protocol only, no results claimed. Source: https://github.com/provael/provael/blob/main/docs/studies/pi0-openpi-transfer.mdFri, 24 Jul 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, no results claimed. Target: SmolVLA × Meta-World — a different task distribution. PRE-REGISTERED — protocol only, no results claimed. Source: https://github.com/provael/provael/blob/main/docs/studies/metaworld-transfer.mdFri, 24 Jul 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, no results claimed. Target: A physical SO-ARM101 arm, supervised table-top trials. PRE-REGISTERED — protocol only, no results claimed. Provael ships no physical-robot control code and no real-world-harm payloads; the physical eval would use LeRobot’s own evaluation loop under direct human supervision with an e-stop. Source: https://github.com/provael/provael/blob/main/docs/studies/sim-to-real-so101.mdFri, 24 Jul 2026 00:00:00 GMThttps://www.provael.com/studies/https://www.provael.com/studies/Protocol filed, since run. Target: Attack transfer, deterministic CPU fixture. Measured on the CPU `reach` fixture only. The SmolVLA × LIBERO row is empty: no real-policy EAI04 transfer is claimed. Source: https://github.com/provael/provael/blob/main/docs/studies/eai04-action-space-transfer.mdTue, 21 Jul 2026 00:00:00 GMT