The Readout, Not the Denoiser: Per-Step Integrated Gradients for Diffusion-Policy Vision-Language-Action Models

Diffusion-based vision-language-action (VLA) policies perform well on robotic manipulation tasks, but their per-decision reasoning remains opaque. To our knowledge, we present the first per-step, per-modality application of Integrated Gradients (IG) to a diffusion-policy VLA. The method attributes every control decision of the Robotics Diffusion Transformer (RDT) to its vision, language, and state inputs across ManiSkill3 manipulation episodes. It defines a deterministic Gaussian log-density target through a fixed-seed denoising chain and runs at the policy-call level to keep full-episode attribution tractable. We evaluate on four tasks with two seeds, covering 1,341 signal-bearing decisions out of roughly 1,900 attributed policy calls. Median completeness error is 1.9 to 2.5% on vision and language. Vision insertion and deletion area-under-curve (AUC) scores of 0.79 to 0.87 and 0.29 to 0.35 pass pre-specified faithfulness bars. Language AUCs miss most of their bars at 170M, where instructions tokenize to only 16 to 30 tokens. On the fine-tuned RDT-1B, language deletion improves to passing and language insertion is seed-sensitive, passing at two of three evaluation seeds. We then report two structural findings about evaluating attribution on diffusion policies. First, the log-density change under deletion does not track policy competence. A fine-tuned RDT-1B improves its vision AUCs at every one of three evaluation seeds (pooled over all policy calls, vision insertion 0.940, deletion 0.288) while its vision deletion magnitude stays near -0.00015 nats. A direct measurement shows this is an artifact of the per-entry quadratic log-density readout, which maps a small action displacement to a near-zero nat value, and not of the action being held fixed by the denoiser. Ranking-based metrics, not magnitudes, are the right faithfulness tool for this setting. Second, model-randomization sanity correlations persist even under frozen-target and full-backbone randomization. This follows from the input-anchored form of IG, while input randomization remains discriminating and passes (rho <= 0.17 on vision). Finally, a raw L2 attribution target restores a usable deletion magnitude of -1.23 at k=5% in action-space units, while the log-density target yields -0.0014 nats on the same population. It keeps AUCs faithful at a quantified cost in completeness. We release the full pipeline code and the committed per-step metrics records. Companion repository containing the code, released metrics records, interval tables, and analysis notebooks: https://github.com/itscool2b/the-readout-not-the-denoiser-repo