LatentQuant: Preserving the Policy-Facing Latent Contract under NVFP4 VAE Quantization

arXiv cs.AIOctober 6, 2026en

arXiv:2610.03959v1 Announce Type: new Abstract: Recent world action models (WAMs) reuse pretrained video VAEs whose encoder latents directly condition downstream action policies. Quantization must therefore preserve not only reconstruction fidelity but also the policy-facing latent contract expected by the frozen policy. Direct NVFP4 leaves W4A4 quantization error uncompensated, whereas joint quantization-aware training (QAT) can recover reconstruction by moving this representation. On Wan2.1, joint QAT nearly matches FP32 VBench-7 (0.7403 versus 0.7409), yet LIBERO success collapses from 95.5% to 10.5%. Controlled decoder-only experiments show that activation quantize-dequantize operations

This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.

• Forskning
• Bild
• Reglering