Datasets:

mosaichunk
/

RememBench

@@ -25,150 +25,36 @@ configs:
 # RememBench
-RememBench evaluates consistency across long-horizon revisits in autoregressive video generation. Target objects or scenes leave the model's sliding window and later return to view. Revisits are driven by prompts in the text-to-video (T2V) split and by camera motion in the image-to-video (I2V) split.
-Introduced in **MosaiChunk: Compositing Spatio-Temporal Memory for Autoregressive Video Generation**.
 [Project page](https://mosaichunk.github.io/) · [Paper](https://mosaichunk.github.io/assets/paper.pdf) · [Video Viewer](https://mosaichunk.github.io/viewer/video.html)
-**Release status:** `0.1.0-rc1`, a private release candidate. The publication license is not yet assigned. This package contains the fixed evaluation inputs, camera trajectories, preparation and validation scripts, and T2V reference frame annotations and scores. DL3DV images, generated videos, model weights, and model inference implementations are not bundled.
-## Dataset contents
-| Subset | Scenes | Evaluation inputs | Input |
 |---|---:|---:|---|
-| `t2v` | 100 | 100 | Four prompt segments, their start times, and a fixed noise seed |
-| `i2v` | 150 | 750 | Scene prompt, initial-frame source, camera trajectory, and a fixed noise seed |
-Both subsets use the `test` split. The I2V set contains **50 indoor and 100 outdoor scenes**. All 150 scenes have 90°, 180°, and 360° rotation trajectories. The 100 outdoor scenes additionally have these three rotations combined with translation. Thus, 750 I2V rows represent 150 unique scenes; they are not 750 independent scenes.
-T2V scenarios are disjoint from router training. I2V conditioning frames come from DL3DV, whereas router training used Sekai frames. The evaluation trajectories were not used during training.
-## Load the inputs
-```bash
-pip install -r requirements.txt
-hf auth login
-```
-While the repository is private, authentication must use an account with access.
 ```python
 from datasets import load_dataset
 t2v = load_dataset("evanbuzzZ/RememBench", "t2v", split="test", token=True)
 i2v = load_dataset("evanbuzzZ/RememBench", "i2v", split="test", token=True)
-sample = t2v[0]
-prompts = sample["prompts"]
-seed = int(sample["seed"])
-```
-**Seeds are decimal strings.** Some T2V seeds exceed the exact-integer range of JavaScript numbers. Convert them directly to an integer in your generation code; do not pass them through a floating-point value.
-Download the repository to use the trajectory files and scripts:
-```bash
-hf download evanbuzzZ/RememBench --repo-type dataset --local-dir RememBench
-cd RememBench
-python scripts/validate.py
-python scripts/load_example.py --subset i2v --index 0
 ```
-For reproducible experiments, pin the Hugging Face repository commit with `revision=` or `--revision`.
-## T2V inputs
-Each row of `data/t2v/test.jsonl` contains:
-| Field | Meaning |
-|---|---|
-| `scene_id`, `source_index` | Stable scene identifier and original candidate index |
-| `target_description` | Target content described by the prompt |
-| `prompts` | Four prompts: appears, disappears, remains out of sight, reappears |
-| `prompt_start_seconds` | Nominal segment starts: `[0.0, 3.6, 6.2, 11.2]` |
-| `seed` | Exact noise seed, stored as a decimal string |
-| `num_frames`, `fps`, `width`, `height` | Paper setting: 379 frames, 24 fps, 1376 × 768 |
-| `num_chunks`, `denoising_steps_per_chunk` | Paper setting: 23 chunks, four denoising steps per chunk |
-The prompt schedule specifies intended action timing, not the exact generated action frames. The five-second third segment keeps the target out of sight longer than the largest sliding-window baseline's approximately three seconds of recent context. The paper retained 100 prompts from a 466-scene candidate pool after screening an H3-AR rollout for compliance. Evaluation uses the same retained prompts and seeds across methods and budgets.
-## I2V inputs and initial frames
-Each row of `data/i2v/test.jsonl` contains:
-| Field | Meaning |
-|---|---|
-| `sample_id`, `scene_id` | Unique scene–trajectory identifier and source DL3DV scene ID |
-| `scene_type` | `indoor` or `outdoor` |
-| `prompt` | Caption describing visible scene content without prescribing camera motion |
-| `image_path` | Relative path populated by the image preparation script |
-| `trajectory_path` | Relative path to a bundled NumPy archive |
-| `setting`, `rotation_degrees`, `direction`, `translation` | Camera-motion setting; direction is `left` or `right` |
-| `seed` | Paper generation seed, `"0"` |
-| `num_frames`, `fps`, `width`, `height` | Paper setting: 253 frames, 16 fps, 832 × 480 |
-| `num_chunks`, `latent_frames`, `denoising_steps_per_chunk` | 16 chunks, 64 latent frames, four denoising steps per chunk |
-| `pose_samples`, `intrinsics_samples` | Lengths of the original camera arrays |
-`sources/i2v_images.jsonl` contains one record per unique scene, with its upstream repository, archive path, retrieval revision, preprocessing rule, and SHA-256 of the expected RGB pixels. Initial frames are taken from the first image member in lexicographic archive-path order, center-cropped to the target aspect ratio, then resized to 832 × 480 using Lanczos.
-Obtain access to [DL3DV-ALL-960P](https://huggingface.co/datasets/DL3DV/DL3DV-ALL-960P) under its own terms. If you already have the scene archives or extracted scene folders:
-```bash
-python scripts/prepare_i2v_images.py --source-dir /path/to/DL3DV-ALL-960P
-python scripts/validate.py --require-images
-```
-Alternatively, retrieve the required first frames through your authorized Hugging Face account:
-```bash
-python scripts/prepare_i2v_images.py --download
-```
-The download option uses HTTP range requests to read the first image from each source archive, without downloading all of its video frames. Use `--limit 1` to try one scene first. Existing prepared images are validated and skipped. The script refuses a pixel-hash mismatch, rather than silently substituting a different input.
-### Camera representation
-Each `.npz` file has `poses` and `intrinsics`, copied without changing their numerical values from the final experiment inputs:
-- `poses`: float32 camera-to-world matrices of shape `(T, 4, 4)`. Camera axes are x right, y down, z forward. The source world frame is z up; the first camera is at `[0, 0, 1.5]`, facing world +y. Translation values retain the source trajectory's depth-derived scale.
-- `intrinsics`: float32 rows `[fx, fy, cx, cy]` in pixels at 832 × 480. Intrinsics are constant over each trajectory. Some source arrays have a different number of rows from `poses`; use the first intrinsics row, as in the paper's implementation.
-The source pose sequence contains 256 or 257 samples, not the 253 decoded video frames. In the paper's LingBot-World-Infinity setup, rotation is interpolated with SLERP and translation linearly at 64 equally spaced positions over the full pose sequence, including both endpoints. The backbone then uses relative consecutive poses with translation normalized by the largest displacement, following its camera-conditioning implementation. Preserve the complete source sequence rather than truncating it to the decoded frame count.
-For 90° and 180° rotation, the camera turns away and reverses. For 360°, it completes a continuous full turn. Translation moves into the scene and returns. All source trajectories end at their initial position and orientation.
-## Evaluation
-The paper uses **RAVEN-adapted MiniMax-H3 (H3-AR)** for T2V and **LingBot-World-Infinity** for I2V. Compare methods using the same prompts and noise seed; I2V methods also share the conditioning frame and input trajectory. Retrieval methods retain two recent chunks and one or two chunks of far memory. Base retains additional recent chunks to match the total active visual KV budget, with the same backbone-specific attention sink.
-T2V departure and revisit frames are manually annotated. The departure frame is shared across methods within a scene; the revisit is marked separately for each generated rollout. I2V uses the conditioning frame as departure and selects each rollout's revisit from its own Pi3X reconstruction. For rotation, find the frame whose viewing direction differs most from the initial direction; for translation, find the frame farthest from the initial position. From that turnaround onward, select the frame with the smallest unsigned viewing-direction angle to the initial frame. Reconstruct all decoded frames, with inputs resized to approximately 255,000 pixels and dimensions rounded to multiples of 14.
-| Metric | Definition | Aggregation |
-|---|---|---|
-| CLIP ↑ | Cosine similarity of L2-normalized CLIP ViT-H/14 LAION-2B embeddings of departure and revisit | Median over scenes |
-| LPIPS ↓ | AlexNet, version 0.1, on native-resolution image pairs scaled to `[-1, 1]` | Median over scenes |
-| TempSSIM ↑ | Grayscale SSIM over consecutive frames, 11 × 11 Gaussian window, sigma 1.5 | Mean within each video, then median over scenes |
-| Drift ↓ | Cosine distance between adjacent chunks, each represented by the normalized average of CLIP embeddings from four evenly spaced frames | Mean within each video, then median over scenes |
-H3-AR decodes an initial five-frame chunk followed by 17-frame chunks; LingBot-World-Infinity decodes an initial 13-frame chunk followed by 16-frame chunks. Scores use original rollouts before compression for the Video Viewer.
-`evaluation/t2v_reference_pairs.jsonl` contains 600 annotations and CLIP/LPIPS measurements: 100 scenes × three methods × two comparison budgets. Frame indices are zero-based. These annotations describe the paper's specific generated rollouts; **they are not reusable ground-truth frame indices for a new model or rerun**. Original rollout videos are not included in this input package. `source_arm` identifies the original run, and `comparison_budget_chunks` identifies the matched-budget comparison; Base has no retrieved far memory.
-`evaluation/t2v_reference_summary.json` reproduces the paper's median T2V CLIP and LPIPS values:
-| Method | CLIP, 1 chunk | LPIPS, 1 chunk | CLIP, 2 chunks | LPIPS, 2 chunks |
-|---|---:|---:|---:|---:|
-| Base | 0.755 | 0.659 | 0.751 | 0.652 |
-| MoC | 0.832 | 0.594 | 0.841 | 0.569 |
-| MosaiChunk | 0.899 | 0.567 | 0.936 | 0.500 |
-## Provenance, limitations, and licensing
-`sources/input_provenance.json` records hashes of the source manifests, prompts, trajectories, seeds, and annotations used for this export. `checksums.sha256` covers the distributed release files.
-Prompt compliance screening used one H3-AR rollout per candidate; success rates can differ with another model. T2V revisit annotation is manual, and I2V revisit selection depends on reconstructed poses. Report the evaluated cohort and any generation failures rather than silently changing the test set. The dataset measures revisits in the supplied scenarios and trajectories; it is not an exhaustive measure of video quality or world understanding.
-Publication licensing for the RememBench-authored material is pending. DL3DV content remains subject to the [DL3DV license and Terms of Use](https://github.com/DL3DV-10K/Dataset/blob/main/License.md); this repository does not grant additional rights to it. The initial-frame pixels are obtained from the upstream source by the user and are not redistributed here. The four-stage T2V design extends the appear–disappear–reappear design of Ring Forcing; construction details and citations are in the linked paper.
-Please cite the MosaiChunk paper when using RememBench and acknowledge DL3DV when using the I2V split. Canonical paper citation metadata will be added with the public release.

 # RememBench
+A benchmark for long-horizon revisit consistency in autoregressive video generation, introduced in **MosaiChunk**.
 [Project page](https://mosaichunk.github.io/) · [Paper](https://mosaichunk.github.io/assets/paper.pdf) · [Video Viewer](https://mosaichunk.github.io/viewer/video.html)
+## Data
+| Subset | Scenes | Inputs | Contents |
 |---|---:|---:|---|
+| `t2v` | 100 | 100 | Four-segment prompts, timing, and seeds |
+| `i2v` | 150 | 750 | Initial-frame sources, scene prompts, camera trajectories, and seeds |
+Both subsets use the `test` split. I2V contains 50 indoor and 100 outdoor scenes. All scenes have 90°, 180°, and 360° rotations; outdoor scenes also have rotation with translation.
+DL3DV initial frames are retrieved using the provided preparation script and are not bundled here.
+## Usage
 ```python
 from datasets import load_dataset
 t2v = load_dataset("evanbuzzZ/RememBench", "t2v", split="test", token=True)
 i2v = load_dataset("evanbuzzZ/RememBench", "i2v", split="test", token=True)
 ```
+Log in with `hf auth login` while the repository is private. Seeds are stored as strings; use `int(sample["seed"])` when generating videos.
+[Download, image preparation, data formats, and evaluation](docs/usage.md)
+CLIP and LPIPS measure departure–revisit consistency. Included T2V frame annotations describe the paper's rollouts; new rollouts require their own revisit annotations.
+## License
+The release license is pending. DL3DV images remain subject to the [DL3DV Terms of Use](https://github.com/DL3DV-10K/Dataset/blob/main/License.md).

checksums.sha256 CHANGED Viewed

@@ -1,8 +1,9 @@
 9e75dd981de037ec3769f24f790e126bc5a160b6871f510214e68dc70649aeeb  .gitattributes
 3065b361e3d3d3f3c511d397cb95433ef8c34bf72b72416b9d7a2ee96b730168  .gitignore
-8d0e4ab89019478b67eb8bd71acdd98559c942ef2defbb3d3e686236a3fcbc11  README.md
 1dc22bccd233317d26e852a8ba148bd84bc93b48449c9ab33eea9f0455510c62  data/i2v/test.jsonl
 9b065b36ed147a7a3a485e00c5dacf453a9c85d0cb62baecb4d9a09bbd2d35f5  data/t2v/test.jsonl
 15a96f50306c85ddc95d197627970ca8c55d6cf4fed049ce16e388e6b9537fb0  evaluation/t2v_reference_pairs.jsonl
 cd543fba727c00e1d84acc8131c71d52d62da0cf4ba1682549298450b333d622  evaluation/t2v_reference_summary.json
 c9a95ee0382d402c4090c0eeb0a197d32ec6ede7d85e803c8d9148708e44a551  release_info.json

 9e75dd981de037ec3769f24f790e126bc5a160b6871f510214e68dc70649aeeb  .gitattributes
 3065b361e3d3d3f3c511d397cb95433ef8c34bf72b72416b9d7a2ee96b730168  .gitignore
+addfdc3f0422c7668200888a93536733408ed9575ded9a0b56c235c1af7b2c86  README.md
 1dc22bccd233317d26e852a8ba148bd84bc93b48449c9ab33eea9f0455510c62  data/i2v/test.jsonl
 9b065b36ed147a7a3a485e00c5dacf453a9c85d0cb62baecb4d9a09bbd2d35f5  data/t2v/test.jsonl
+10c89128eaccaefcc98762c386b2eff048e941dd95b6fd07b33c452a2f49cdd6  docs/usage.md
 15a96f50306c85ddc95d197627970ca8c55d6cf4fed049ce16e388e6b9537fb0  evaluation/t2v_reference_pairs.jsonl
 cd543fba727c00e1d84acc8131c71d52d62da0cf4ba1682549298450b333d622  evaluation/t2v_reference_summary.json
 c9a95ee0382d402c4090c0eeb0a197d32ec6ede7d85e803c8d9148708e44a551  release_info.json

docs/usage.md ADDED Viewed

	@@ -0,0 +1,133 @@

+# RememBench usage guide
+[Dataset overview](../README.md)
+## Download and load
+```bash
+pip install huggingface_hub
+hf auth login
+hf download evanbuzzZ/RememBench --repo-type dataset --local-dir RememBench
+cd RememBench
+pip install -r requirements.txt
+```
+While the repository is private, authentication must use an account with access.
+```python
+from datasets import load_dataset
+t2v = load_dataset("evanbuzzZ/RememBench", "t2v", split="test", token=True)
+i2v = load_dataset("evanbuzzZ/RememBench", "i2v", split="test", token=True)
+sample = t2v[0]
+prompts = sample["prompts"]
+seed = int(sample["seed"])
+```
+**Seeds are decimal strings.** Some T2V seeds exceed the exact-integer range of JavaScript numbers. Convert them directly to an integer in your generation code; do not pass them through a floating-point value.
+Check the downloaded data and inspect one sample:
+```bash
+python scripts/validate.py
+python scripts/load_example.py --subset i2v --index 0
+```
+For reproducible experiments, pin the Hugging Face repository commit with `revision=` or `--revision`.
+## T2V inputs
+Each row of `data/t2v/test.jsonl` contains:
+| Field | Meaning |
+|---|---|
+| `scene_id`, `source_index` | Stable scene identifier and original candidate index |
+| `target_description` | Target content described by the prompt |
+| `prompts` | Four prompts: appears, disappears, remains out of sight, reappears |
+| `prompt_start_seconds` | Nominal segment starts: `[0.0, 3.6, 6.2, 11.2]` |
+| `seed` | Exact noise seed, stored as a decimal string |
+| `num_frames`, `fps`, `width`, `height` | Paper setting: 379 frames, 24 fps, 1376 × 768 |
+| `num_chunks`, `denoising_steps_per_chunk` | Paper setting: 23 chunks, four denoising steps per chunk |
+The prompt schedule specifies intended action timing, not the exact generated action frames. The five-second third segment keeps the target out of sight longer than the largest sliding-window baseline's approximately three seconds of recent context. The paper retained 100 prompts from a 466-scene candidate pool after screening an H3-AR rollout for compliance. Evaluation uses the same retained prompts and seeds across methods and budgets.
+## I2V inputs and initial frames
+Each row of `data/i2v/test.jsonl` contains:
+| Field | Meaning |
+|---|---|
+| `sample_id`, `scene_id` | Unique scene–trajectory identifier and source DL3DV scene ID |
+| `scene_type` | `indoor` or `outdoor` |
+| `prompt` | Caption describing visible scene content without prescribing camera motion |
+| `image_path` | Relative path populated by the image preparation script |
+| `trajectory_path` | Relative path to a bundled NumPy archive |
+| `setting`, `rotation_degrees`, `direction`, `translation` | Camera-motion setting; direction is `left` or `right` |
+| `seed` | Paper generation seed, `"0"` |
+| `num_frames`, `fps`, `width`, `height` | Paper setting: 253 frames, 16 fps, 832 × 480 |
+| `num_chunks`, `latent_frames`, `denoising_steps_per_chunk` | 16 chunks, 64 latent frames, four denoising steps per chunk |
+| `pose_samples`, `intrinsics_samples` | Lengths of the original camera arrays |
+`sources/i2v_images.jsonl` contains one record per unique scene, with its upstream repository, archive path, retrieval revision, preprocessing rule, and SHA-256 of the expected RGB pixels. Initial frames are taken from the first image member in lexicographic archive-path order, center-cropped to the target aspect ratio, then resized to 832 × 480 using Lanczos.
+Obtain access to [DL3DV-ALL-960P](https://huggingface.co/datasets/DL3DV/DL3DV-ALL-960P) under its own terms. If you already have the scene archives or extracted scene folders:
+```bash
+python scripts/prepare_i2v_images.py --source-dir /path/to/DL3DV-ALL-960P
+python scripts/validate.py --require-images
+```
+Alternatively, retrieve the required first frames through your authorized Hugging Face account:
+```bash
+python scripts/prepare_i2v_images.py --download
+```
+The download option uses HTTP range requests to read the first image from each source archive, without downloading all of its video frames. Use `--limit 1` to try one scene first. Existing prepared images are validated and skipped. The script refuses a pixel-hash mismatch, rather than silently substituting a different input.
+### Camera representation
+Each `.npz` file has `poses` and `intrinsics`, copied without changing their numerical values from the final experiment inputs:
+- `poses`: float32 camera-to-world matrices of shape `(T, 4, 4)`. Camera axes are x right, y down, z forward. The source world frame is z up; the first camera is at `[0, 0, 1.5]`, facing world +y. Translation values retain the source trajectory's depth-derived scale.
+- `intrinsics`: float32 rows `[fx, fy, cx, cy]` in pixels at 832 × 480. Intrinsics are constant over each trajectory. Some source arrays have a different number of rows from `poses`; use the first intrinsics row, as in the paper's implementation.
+The source pose sequence contains 256 or 257 samples, not the 253 decoded video frames. In the paper's LingBot-World-Infinity setup, rotation is interpolated with SLERP and translation linearly at 64 equally spaced positions over the full pose sequence, including both endpoints. The backbone then uses relative consecutive poses with translation normalized by the largest displacement, following its camera-conditioning implementation. Preserve the complete source sequence rather than truncating it to the decoded frame count.
+For 90° and 180° rotation, the camera turns away and reverses. For 360°, it completes a continuous full turn. Translation moves into the scene and returns. All source trajectories end at their initial position and orientation.
+## Evaluation
+The paper uses **RAVEN-adapted MiniMax-H3 (H3-AR)** for T2V and **LingBot-World-Infinity** for I2V. Compare methods using the same prompts and noise seed; I2V methods also share the conditioning frame and input trajectory. Retrieval methods retain two recent chunks and one or two chunks of far memory. Base retains additional recent chunks to match the total active visual KV budget, with the same backbone-specific attention sink.
+T2V departure and revisit frames are manually annotated. The departure frame is shared across methods within a scene; the revisit is marked separately for each generated rollout. I2V uses the conditioning frame as departure and selects each rollout's revisit from its own Pi3X reconstruction. For rotation, find the frame whose viewing direction differs most from the initial direction; for translation, find the frame farthest from the initial position. From that turnaround onward, select the frame with the smallest unsigned viewing-direction angle to the initial frame. Reconstruct all decoded frames, with inputs resized to approximately 255,000 pixels and dimensions rounded to multiples of 14.
+| Metric | Definition | Aggregation |
+|---|---|---|
+| CLIP ↑ | Cosine similarity of L2-normalized CLIP ViT-H/14 LAION-2B embeddings of departure and revisit | Median over scenes |
+| LPIPS ↓ | AlexNet, version 0.1, on native-resolution image pairs scaled to `[-1, 1]` | Median over scenes |
+| TempSSIM ↑ | Grayscale SSIM over consecutive frames, 11 × 11 Gaussian window, sigma 1.5 | Mean within each video, then median over scenes |
+| Drift ↓ | Cosine distance between adjacent chunks, each represented by the normalized average of CLIP embeddings from four evenly spaced frames | Mean within each video, then median over scenes |
+H3-AR decodes an initial five-frame chunk followed by 17-frame chunks; LingBot-World-Infinity decodes an initial 13-frame chunk followed by 16-frame chunks. Scores use original rollouts before compression for the Video Viewer.
+`evaluation/t2v_reference_pairs.jsonl` contains 600 annotations and CLIP/LPIPS measurements: 100 scenes × three methods × two comparison budgets. Frame indices are zero-based. These annotations describe the paper's specific generated rollouts; **they are not reusable ground-truth frame indices for a new model or rerun**. Original rollout videos are not included in this input package. `source_arm` identifies the original run, and `comparison_budget_chunks` identifies the matched-budget comparison; Base has no retrieved far memory.
+`evaluation/t2v_reference_summary.json` reproduces the paper's median T2V CLIP and LPIPS values:
+| Method | CLIP, 1 chunk | LPIPS, 1 chunk | CLIP, 2 chunks | LPIPS, 2 chunks |
+|---|---:|---:|---:|---:|
+| Base | 0.755 | 0.659 | 0.751 | 0.652 |
+| MoC | 0.832 | 0.594 | 0.841 | 0.569 |
+| MosaiChunk | 0.899 | 0.567 | 0.936 | 0.500 |
+## Provenance, limitations, and licensing
+`sources/input_provenance.json` records hashes of the source manifests, prompts, trajectories, seeds, and annotations used for this export. `checksums.sha256` covers the distributed release files.
+Prompt compliance screening used one H3-AR rollout per candidate; success rates can differ with another model. T2V revisit annotation is manual, and I2V revisit selection depends on reconstructed poses. Report the evaluated cohort and any generation failures rather than silently changing the test set. The dataset measures revisits in the supplied scenarios and trajectories; it is not an exhaustive measure of video quality or world understanding.
+Publication licensing for the RememBench-authored material is pending. DL3DV content remains subject to the [DL3DV license and Terms of Use](https://github.com/DL3DV-10K/Dataset/blob/main/License.md); this repository does not grant additional rights to it. The initial-frame pixels are obtained from the upstream source by the user and are not redistributed here. The four-stage T2V design extends the appear–disappear–reappear design of Ring Forcing; construction details and citations are in the linked paper.
+Please cite the MosaiChunk paper when using RememBench and acknowledge DL3DV when using the I2V split. Canonical paper citation metadata will be added with the public release.