Searcharxiv⌕ Search

arXiv subjects

Amit Jadhav

Publications and source records attributed to Amit Jadhav.

4 recordsLinked to original sources

Brain-Aligned Latent Actions: Imposing Premotor Cortex Geometry on a Robot's Action Vocabulary

Robot policies increasingly act through a small vocabulary of discrete latent actions learned from unlabeled video, yet training leaves open which motion receives which code. We find this choice highly unstable: two independently trained runs of Moto's latent motion tokenizer assign the same robot clips to nearly unrelated codes (adjusted Rand index 0.001), and evaluating one tokenizer in bf16 instead of fp32 reassigns 74% of its tokens. We use this freedom to impose the representational geometry of macaque ventral premotor cortex (PMv), recorded while a monkey drew shapes, on the tokenizer's code assignment, with a straight-through similarity loss trained alongside its reconstruction objective on robot video. In one training run per target, on five shape categories the loss never saw, the PMv-trained tokenizer reaches a representational similarity of 0.52 with PMv, against 0.20-0.26 for released and no-prior tokenizers and 0.42 for the 95th percentile of random kinematic encoders. Kinematic, language, and random targets do not transfer (-0.17 to -0.03). Two results limit the claim: a hand-picked kinematic descriptor, the stroke's initial direction, matches PMv better (0.65), and a label-shuffled PMv control scores 0.41. We specify the downstream manipulation test, on a CALVIN pipeline that reproduces Moto's published results, before running it.

cs.RO↗

Streaming Failure Forecasting: VLA Failure Monitors That Keep Learning After Deployment

Failure monitors for vision-language-action (VLA) policies are trained offline and then frozen. After deployment, lighting, objects and cameras drift, and the policy itself is updated or moved to new tasks. On recorded real-robot runs, after a switch of task or policy version, a frozen monitor fit on the old task ranks failures little better than chance. We present Streaming Failure Forecasting (SFF), a monitor that keeps learning from the one label every episode provides: whether it succeeded. SFF keeps the offline monitor's network fixed and refits only its last linear layer after each episode, at 2.18 ms per policy call on a Jetson CPU core. Its alarm threshold moves only after successes, which in simulation kept false alarms near target except when two drifts came at once. With every setting locked before testing, SFF ranks failures better than the same monitor kept frozen on 4 of 5 simulated test suites under unannounced drift, and above resetting on detected drift on every suite. Carried with its locked learner settings to 434 recorded runs of four VLAs on real SO-101 arms, replayed across unannounced switches of task or policy version, it ranks failures better than the frozen monitor for every VLA and, in an exploratory comparison, above a full reset at the true switch. A deployed failure monitor need not stay frozen: refitting its readout from outcomes is cheap and helps after the changes we tested.

cs.RO↗

Undertone: Personalizing Frozen VLM Aesthetic Judges from Few Ratings

Image quality assessment predicts a mean opinion score (MOS), but for image aesthetics the mean hides the individual, so personalized assessment must start from the few images a new user rates. Fitting each person from scratch fails with ten ratings; networks fine-tuned per person carry over poorly to new kinds of images. We present Undertone-P, which learns from other raters a Gaussian prior over per-person regression weights on read-outs of frozen vision-language model (VLM) judges: the probability-weighted rating and the hidden state at the score token. Since people differ from the crowd in ways other raters reveal, a few ratings need only choose among those ways: a new user costs one closed-form solve, and no network is trained. On Ryu and Yanaka's PARA benchmark, the prior alone reaches a Spearman correlation of .640, above the best published 10- and 100-rating results (.581 and .620); 10 and 100 of a person's ratings raise it to .648 and .676. These ratings add most where tastes depart from the crowd: with 100 ratings, +.067 for the raters who agree least with it and +.206 on LAPIS artworks, against +.037 overall. Without labels, the judges' expected scores predict PARA's MOS at .835 (best published model we ran: .694). PARA-trained networks still lead under PARA's original protocol, likely through a stronger end-to-end MOS model.

cs.CV↗

Electrospun nanodiamond-silk fibroin membranes: a multifunctional platform for biosensing and wound healing applications

Next generation wound care technology capable of diagnosing wound parameters, promoting healthy cell growth and reducing pathogenic infections noninvasively will provide patients with an improved standard of care and an accelerated wound repair. Temperature is one of the indicating biomarkers specific to chronic wounds. This work reports a hybrid, multifunctional optical platform: nanodiamond-silk membranes as bioinspired dressings capable of temperature sensing and wound healing. The hybrid was fabricated through electrospinning and formed sub-micron fibrous membranes with high porosity. The silk fibres are capable of compensating for the lack of extracellular matrix at the wound site, supporting the wound healing. The negatively charged nitrogen vacancy (NV-) color centres in nanodiamonds (NDs) exhibit optically detected magnetic resonance (ODMR) properties and act as fluorescent nanoscale thermometers, capable of sensing temperature variations associated to the presence of infection or inflammation in a wound, without physically removing the dressing. Our results show that the presence of NDs in the hybrid ND-silk membranes improve the thermal stability of silk fibres. The NV- color centres in NDs embedded in silk fibres exhibit well-retained fluorescent and ODMR. Using the NV- centres as fluorescent nanoscale thermometers, we achieved temperature sensing at a range of temperatures, including the biologically relevant temperature window, on cell-cultured ND-silk membranes. Enhancement in the temperature sensitivity of the NV- centres was observed for the hybrids. The membranes were further tested in vivo in a murine wound healing model and demonstrated biocompatibility and equivalent wound closure rates as the control wounds. Additionally, the hybrid ND-silk membranes showed selective antifouling and biocidal propensity toward Gram-negative Pseudomonas aeruginosa and Escherichia coli.

physics.app-ph↗