docs: add RBench results and prompt enhancer

README.md

@@ -29,7 +29,7 @@ tags: [kandinsky, image-to-video, physical-ai, world-model]
 - [Overview](#overview)
 - [Model Zoo](#model-zoo)
 - [Examples](#examples) — [🚗 Autonomous Driving](#🚗-autonomous-driving) · [🤖 Robotics](#🤖-robotics) · [🌍 General Physics](#🌍-general-physics)
-- [Results](#results) — [PAI-Bench-G](#pai-bench-g-physical-ai-bench--generation) · [Physics-IQ Verified](#physics-iq-verified)
+- [Results](#results) — [PAI-Bench-G](#pai-bench-g-physical-ai-bench--generation) · [RBench](#rbench) · [Physics-IQ Verified](#physics-iq-verified)
 - [Quickstart](#quickstart) — [Path A: GitHub code](#path-a--github-code) · [Path B: Diffusers](#path-b--diffusers)
 - [Acknowledgements](#acknowledgements)
 
@@ -113,9 +113,9 @@ First frames + prompts: [`assets/general/`](https://github.com/kandinskylab/kand
 
 ## Results
 
-**Kandinsky WM 1.0** on two physical-AI video benchmarks (our row in **bold**; rows copied as-is
-from our internal leaderboard). Physics-IQ was run with a Qwen3-VL prompt enhancer — noted above the
-table, with the script under [`prompt_enhancers/`](https://github.com/kandinskylab/kandinsky-wm/tree/main/prompt_enhancers). Sizes are total parameter
+**Kandinsky WM 1.0** on three physical-AI video benchmarks (our row in **bold**). RBench and
+Physics-IQ were run with Qwen3-VL prompt enhancers — noted above each table, with the scripts under
+[`prompt_enhancers/`](https://github.com/kandinskylab/kandinsky-wm/tree/main/prompt_enhancers). Sizes are total parameter
 counts where publicly disclosed (MoE models note active params; proprietary/undisclosed left as —).
 
 ### PAI-Bench-G (Physical AI Bench — Generation)
@@ -141,6 +141,36 @@ Physical-AI post-training lifts the base **K5 Lite** by **+1.2 Overall** (80.5
 **+3.0 on the Domain axis** (83.0 → 86.0) — a 2B model landing between Veo-3 and the
 Cosmos-Predict2.5 family.
 
+### RBench
+
+[Leaderboard](https://huggingface.co/spaces/DAGroup-PKU/RBench-Leaderboard), **Qwen** evaluator tab.
+RBench evaluates robot-oriented image-to-video generation across five task categories and four
+robot embodiments.
+
+> **Prompt enhancer used:** [`prompt_enhancers/rbench/enhance_rbench_prompts_qwen_v3.py`](https://github.com/kandinskylab/kandinsky-wm/blob/main/prompt_enhancers/rbench/enhance_rbench_prompts_qwen_v3.py) — conservative Qwen3-VL image-grounded action canonicalization.
+
+| Rank | Model | Size | Avg. | Common Manipulation | Spatial Relationship | Multi-entity Collaboration | Long-horizon Planning | Visual Reasoning | Single Arm | Dual Arm | Quadruped Robot | Humanoid Robot |
+|---|---|---|---|---|---|---|---|---|---|---|---|---|
+| 1 | Veo 3 | — | 0.784 | 0.897 | 0.740 | 0.924 | 0.854 | 0.750 | 0.742 | 0.708 | 0.726 | 0.716 |
+| 2 | Wan 2.5 | — | 0.781 | 0.888 | 0.825 | 0.916 | 0.719 | 0.738 | 0.742 | 0.744 | 0.711 | 0.742 |
+| 3 | Hailuo v2 | — | 0.762 | 0.843 | 0.840 | 0.892 | 0.705 | 0.820 | 0.696 | 0.706 | 0.632 | 0.720 |
+| 4 | Seedance 1.0 | — | 0.755 | 0.856 | 0.665 | 0.899 | 0.716 | 0.790 | 0.708 | 0.730 | 0.674 | 0.759 |
+| 5 | Wan2.2_A14B | 27B (14B active) | 0.698 | 0.709 | 0.660 | 0.921 | 0.681 | 0.550 | 0.688 | 0.678 | 0.670 | 0.728 |
+| 6 | **Kandinsky-WM-1.0** | **2B** | **0.651** | **0.836** | **0.685** | **0.846** | **0.657** | **0.524** | **0.509** | **0.568** | **0.578** | **0.656** |
+| 7 | Cosmos 2.5 | 14B | 0.632 | 0.688 | 0.512 | 0.768 | 0.597 | 0.507 | 0.646 | 0.647 | 0.639 | 0.688 |
+| 8 | LongCat-Video | 13.6B | 0.609 | 0.678 | 0.465 | 0.814 | 0.490 | 0.354 | 0.698 | 0.602 | 0.666 | 0.710 |
+| 9 | DreamGen(gr1) | 14B | 0.575 | 0.507 | 0.500 | 0.848 | 0.353 | 0.405 | 0.660 | 0.632 | 0.611 | 0.656 |
+| 10 | Wan2.2_5B | 5B | 0.551 | 0.598 | 0.402 | 0.722 | 0.450 | 0.420 | 0.511 | 0.534 | 0.638 | 0.682 |
+| 11 | Wan2.1_14B | 14B | 0.542 | 0.688 | 0.400 | 0.702 | 0.465 | 0.270 | 0.519 | 0.562 | 0.604 | 0.664 |
+| 12 | SkyReels | 13B | 0.531 | 0.546 | 0.400 | 0.687 | 0.324 | 0.358 | 0.612 | 0.574 | 0.654 | 0.628 |
+| 13 | DreamGen(droid) | 14B | 0.514 | 0.465 | 0.505 | 0.591 | 0.302 | 0.386 | 0.589 | 0.570 | 0.584 | 0.633 |
+| 14 | LTX-Video | 2B | 0.486 | 0.450 | 0.382 | 0.734 | 0.358 | 0.286 | 0.487 | 0.487 | 0.588 | 0.603 |
+| 15 | FramePack | 13B | 0.453 | 0.455 | 0.240 | 0.630 | 0.197 | 0.345 | 0.403 | 0.500 | 0.670 | 0.635 |
+| 16 | CogVideoX_5B | 5B | 0.322 | 0.290 | 0.240 | 0.426 | 0.096 | 0.030 | 0.374 | 0.422 | 0.494 | 0.524 |
+| 17 | Vidar | — | 0.207 | 0.118 | 0.140 | 0.082 | 0.019 | 0.030 | 0.344 | 0.390 | 0.380 | 0.364 |
+| 18 | UnifoLM-WMA-0 | — | 0.104 | 0.029 | 0.065 | 0.025 | 0.000 | 0.000 | 0.290 | 0.106 | 0.251 | 0.170 |
+
+### Physics-IQ Verified
 ### Physics-IQ Verified
 
 > **Prompt enhancer used:** [`prompt_enhancers/physics_iq/temporal_expansion.py`](https://github.com/kandinskylab/kandinsky-wm/blob/main/prompt_enhancers/physics_iq/temporal_expansion.py) — Qwen3-VL temporal caption expansion.
@@ -231,5 +261,6 @@ Built on [Kandinsky 5.0](https://github.com/kandinskylab/kandinsky-5)
 ([HunyuanVideo VAE](https://huggingface.co/hunyuanvideo-community/HunyuanVideo),
 [Qwen2.5-VL](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct),
 [CLIP](https://huggingface.co/openai/clip-vit-large-patch14)). Evaluated on
-[PAI-Bench-G](https://huggingface.co/spaces/shi-labs/physical-ai-bench-leaderboard) and
-[Physics-IQ](https://github.com/google-deepmind/physics-IQ-benchmark).
+[PAI-Bench-G](https://huggingface.co/spaces/shi-labs/physical-ai-bench-leaderboard),
+[Physics-IQ](https://github.com/google-deepmind/physics-IQ-benchmark), and
+[RBench](https://huggingface.co/spaces/DAGroup-PKU/RBench-Leaderboard).