CARE-Drive: Coupled Alternating Refinement for Ego-trajectories

Final inference checkpoints for CARE-Drive, accompanying the training and inference code.

CARE-Drive combines a predictive Qwen3-VL backbone with a waypoint residual expert and a position-conditioned heading expert. The position expert corrects eight future waypoints with bounded metric residuals; the heading expert reconstructs heading from eight adjacent angular increments conditioned on the corrected path.

Released weights

Component	Stage	File
Predictive backbone	A0-DYNAMIC-L1, epoch 5	A0-DYNAMIC-L1/epoch_005.pt
Position expert	C3-XY-EP1, epoch 5	C3-XY-EP1/epoch_005.pt
Final heading expert	D1-H-ADAPT, epoch 5	D1-H-ADAPT/epoch_005.pt

These are the final model tensors used for the paper's 88.837 PDMS system on 12,146 NAVSIM v1 navtest scenes. The score uses the original official CPU evaluation. The release extracts model weights from the accepted training checkpoints, preserving every tensor's value, name, shape, and dtype. Optimizer, scheduler, RNG, and internal experiment metadata are omitted. This is an inference/model-initialization release rather than an optimizer-resume snapshot.

checkpoint_manifest.json records the exported file sizes and SHA256 checksums, original checkpoint checksums, and matching tensor digests. Backbone tensors retain their original BF16/FP32 dtypes; both experts retain FP32. The two experts have 78,420,113 and 78,418,577 parameters, respectively.

Download and run

Install the CARE-Drive environment, then run from the code repository root:

hf download maxiao123/CARE-Drive \
  --include "A0-DYNAMIC-L1/*" "C3-XY-EP1/*" "D1-H-ADAPT/*" "checkpoint_manifest.json" \
  --local-dir runs/care-drive

The current loader first constructs the backbone from the original Qwen model directory and then loads the trained CARE-Drive state. Download the base model and set model.vlm_path accordingly:

export CARE_DRIVE_DATA=/absolute/path/to/data
hf download Qwen/Qwen3-VL-2B-Instruct \
  --local-dir "$CARE_DRIVE_DATA/models/Qwen3-VL-2B-Instruct"

Edit configs/care_drive.yaml for your device and paths. Prepare an observation-only manifest using the data guide, then use the inference command:

export CARE_DRIVE_PERFORMANCE_PREFLIGHT=/absolute/path/to/performance/inference.json
python -m care_drive.infer \
  --config configs/care_drive.yaml \
  --manifest "$CARE_DRIVE_DATA/manifests/navtest.jsonl" \
  --sensor-root "$CARE_DRIVE_DATA/sensor_blobs/test" \
  --backbone runs/care-drive/A0-DYNAMIC-L1/epoch_005.pt \
  --position runs/care-drive/C3-XY-EP1/epoch_005.pt \
  --heading runs/care-drive/D1-H-ADAPT/epoch_005.pt \
  --batch-size 1 --output runs/care-drive/navtest_predictions.jsonl

Formal batch execution uses the codebase's performance setup. V-JEPA2.1 supplies training targets and is not needed for inference.

Each input contains four historical frames from three cameras, ego pose history, velocity, acceleration, and a route command. Each output is eight (x, y, heading) poses at 0.5-second intervals, covering four seconds. Coordinates use the current ego rear-axle frame, with meters for position and radians for heading.

Provenance and validation

The backbone was trained on 92,772 scenes; the expert chain used 92,614 scoreable scenes. The released configuration records the paper's quantization and architecture. Exact cohort manifests are not included.

Export verification compares the complete tensor content against the original accepted checkpoints. All three checkpoints passed strict loading with the published CARE-Drive implementation, including the backbone tokenizer contract. Loaded tensor values and dtypes match the exports exactly, and a single synthetic feature sample passed coupled expert generation. The paper's benchmark result comes from the original evaluation; the newly packaged entry points have not been used to rerun full navtest. See the code repository's validation record for the checked scope.

The model is intended for research in driving trajectory planning. Data, maps, and the original Qwen model assets are obtained separately from their upstream projects. This repository distributes CARE-Drive's trained weights under Apache-2.0; upstream dependencies retain their respective licenses.