SueMintony/CLAMP-HMM-Checkpoints

CLAMP HMM Checkpoints

Neural-HMM priors trained on Llama-3.1 continuation samples for two EAI datasets (BEHAVIOR and VirtualHome). Used by eai_ctrlg as the semantic prior for constrained decoding with γ+β DFAs.

Files

• behavior/hmm-h128-lr0.01/checkpoint.eqx — hidden=128, lr=0.01
• virtualhome/hmm-h128-lr0.01/checkpoint.eqx — hidden=128, lr=0.01

Loading

import equinox as eqx
from ctrlg.hmm.model import ConditionalHMM
model = eqx.tree_deserialise_leaves(
    "behavior/hmm-h128-lr0.01/checkpoint.eqx", ConditionalHMM(...))

Training pipeline: see eai_train/cond_hmm/ in the companion repo.

Experimental checkpoint: IV35-H2-S5 (InternVL3.5)

experimental/iv35-h2-s5-internvl3.5-8b/hmm_formal/checkpoint.eqx is an experimental neural-HMM prior trained from frozen OpenGVLab/InternVL3_5-8B continuations. It is not an InternVL base-model checkpoint and does not replace the two stable BEHAVIOR and VirtualHome checkpoints above.

• Training status: PASS; independent 50-epoch formal run
• Training / validation sequences: 13,170 / 1,460
• Best validation loss: 2.235296
• SHA-256: 3c5f90c6ed76879ba465f7dd009e62a9c87a62905e8f77b8dfffe428527d2512
• Reproducibility metadata: experimental/iv35-h2-s5-internvl3.5-8b/hmm_formal/release_metadata.json

The checkpoint passed its recorded training and reload validation gates. No downstream task-gain claim is made for this experimental release.

Related data and code

• Code: HLR/CLAMP
• VLABench demonstrations: SueMintony/CLAMP-Sampled-Continuations-and-Demos
• Release collection: CLAMP — HMM Checkpoints, Sampled Continuations, and Demos

Published training artifacts

The full, machine-readable pairing of every checkpoint and sampled-continuation artifact is in release_catalog.json.

• Qwen3.5-9B: 10 archived HMM checkpoints under trained/qwen3.5-9b/ — combined BEHAVIOR and VirtualHome variants plus four task-specific variants per domain. Their continuation files are in the Dataset repository.
• InternVL3.5-8B: IV35-H2-S5 formal EQX checkpoint, portable NPZ export, and 2,926 sampled records.
• Qwen3-VL-8B-Instruct: compatibility-remapped HMM. It is a vocabulary remap of a historical HMM, not a retrained model, and has no local sampled-continuation artifact.

The public continuation release contains token arrays / sampled outputs and length metadata only; it excludes original prompts, scene assets, and benchmark data.