Table 5: All task variations except shape used in VLMbench. The shape variation of each task can be found in the detail descriptions of each task category. Variations Totals Values Color 25 seen:red, maroon, lime, green, blue,navy, yellow, cyan, magenta, silver, gray, olive, purple, teal, azure, violet, rose, black, white unseen: brown, gold, pink, chocolate, coral Size 5 larger, smaller, large, medium, small Relative Position 5 top, front, rear, left, right Level 3 top, middle, bottom Amount 2 fully, slightly Action Type 2 open, close Table 6: All object models used in VLMbench. The number behind the object class indicate the instance number of that class. Here, we list variations used for these tasks in Table. 5. For each demonstration, all things in the scene will change the pose at the beginning. When building an instance-level task with one variation, the other variations will also randomly change. For example, in the demonstrations of "Pick & Place objects" with "size" variation, all objects' color and relative positions, including targets and distractors, will randomly change. In the dataset, we have five types of objects, shown in Table 6. We will explain each task in detail as follows. Visualizations can be found on the project website. A.1 Pick & Place Objects Task Definition: The agent needs to distinguish the specific object to grasp and then place it into a particular container. The object can be placed anywhere with any orientation inside the container.

Score distillation sampling (SDS) has shown great promise in text-to-3D generation by distilling pretrained large-scale text-to-image diffusion models, but suffers from over-saturation, over-smoothing, and low-diversity problems. In this work, we propose to model the 3D parameter as a random variable instead of a constant as in SDS and present variational score distillation (VSD), a principled particle-based variational framework to explain and address the aforementioned issues in text-to-3D generation. We show that SDS is a special case of VSD and leads to poor samples with both small and large CFG weights. In comparison, VSD works well with various CFG weights as ancestral sampling from diffusion models and simultaneously improves the diversity and sample quality with a common CFG weight (i.e., $7.5$). We further present various improvements in the design space for text-to-3D such as distillation time schedule and density initialization, which are orthogonal to the distillation algorithm yet not well explored. Our overall approach, dubbed ProlificDreamer, can generate high rendering resolution (i.e., $512\times512$) and high-fidelity NeRF with rich structure and complex effects (e.g., smoke and drops). Further, initialized from NeRF, meshes fine-tuned by VSD are meticulously detailed and photo-realistic. Project page and codes: https://ml.cs.tsinghua.edu.cn/prolificdreamer/

Benefiting from language flexibility and compositionality, humans naturally intend to use language to command an embodied agent for complex tasks such as navigation and object manipulation. In this work, we aim to fill the blank of the last mile of embodied agents -- object manipulation by following human guidance, e.g., "move the red mug next to the box while keeping it upright." To this end, we introduce an Automatic Manipulation Solver (AMSolver) system and build a Vision-and-Language Manipulation benchmark (VLMbench) based on it, containing various language instructions on categorized robotic manipulation tasks. Specifically, modular rule-based task templates are created to automatically generate robot demonstrations with language instructions, consisting of diverse object shapes and appearances, action types, and motion constraints. We also develop a keypoint-based model 6D-CLIPort to deal with multi-view observations and language input and output a sequence of 6 degrees of freedom (DoF) actions. We hope the new simulator and benchmark will facilitate future research on language-guided robotic manipulation.