arXiv:2608. 05025v2 Announce Type: replace Abstract: Joint Energy-Based Models (JEM) unify classification and generation within a single network and support out-of-distribution (OOD) detection.
arXiv:2608. 05025v1 Announce Type: new Abstract: Joint Energy-Based Models (JEM) unify classification and generation within a single network and support out-of-distribution (OOD) detection.
FORGE is a forward‑only test‑time adaptation technique designed for integer‑only vision models running on microcontrollers. It restores batch‑normalization statistics after BN folding by re‑normalizing each convolution’s per‑channel output using only forward‑pass estimates, enabling adaptation on deployed, folded integer models. The method achieves accuracy gains comparable to gradient‑based TENT, requires adapting only a few layers, works with single‑sample streaming, and has been validated on an ESP32‑S3 with minimal energy and latency overhead.
The paper introduces a schema‑adaptive action‑conditioned Joint‑Embedding Predictive Architecture (SAAC‑JEPA) for cross‑machine CNC transfer when only a subset of sensors overlap between source and target machines. Experiments show that pretraining does not improve source‑only forecasting, but a carefully selected action‑conditioned JEPA model achieves a zero‑shot RMSE of 0.546 on the target, outperforming persistence but falling short of certain baseline models. Ablation studies reveal that adding RevIN improves RMSE but harms calibration, and limited post‑lock adaptation can further reduce error.
The paper introduces the Shift‑Invariant Variance Estimator (SIVE) to remove same‑state noise from thermodynamic variance calculations in neural‑network training. SIVE provides an unbiased estimate of path variance without requiring Markov chain Monte Carlo stationarity, and it reveals a reproducible mid‑training trough and rebound in MNIST MLP trajectories that raw variance masks. Experiments across multiple localization scales confirm that SIVE captures a large portion of observation noise and satisfies an early‑drop/late‑rise criterion in most trajectory‑scale pairs.
The paper investigates when online adaptation benefits edge time‑series forecasting under distribution drift, using a leakage‑free streaming protocol on six public multivariate datasets. It shows that the warmup budget for static baselines and the choice of learning rate can bias perceived adaptation gains, and that a validation‑only procedure selecting warmup and optimizer rates yields Adam outperforming SGD with momentum in most settings. The study also examines accuracy versus adaptation‑state memory and per‑update latency for different adaptation strategies, highlighting parameter‑efficient variants that are nondominated on the memory axis.
arXiv:2607. 24507v1 Announce Type: cross Abstract: Existing methods mainly adapt pretrained autoregressive (AR) language models to masked diffusion, whereas we directly adapt them to uniform-noise diffusion, where every token remains editable during sampling.
arXiv:2607. 00958v1 Announce Type: new Abstract: Time series are central to modern data mining applications, from industrial telemetry and server metrics to finance and physiology, yet time-series self-supervised learning often depends on view and augmentation choices that encode domain-specific invariances.
The paper introduces CASTER, a gradient‑free test‑time adaptation method that keeps the model frozen by storing source class statistics in a discriminative subspace and applying an affine transformation estimated from target‑batch moments. CASTER avoids backward passes, optimizer state, and large feature banks, outperforming k‑NN on frozen features in most backbone‑dataset settings while using far less memory. The authors also propose a residual‑to‑margin transportability certificate that flags when affine transport is unreliable, and demonstrate that gating based on this certificate can recover performance losses.
arXiv:2603. 14717v2 Announce Type: replace Abstract: Generating novel protein sequences that respect a family's statistical constraints typically requires training deep generative models on thousands to millions of examples.
arXiv:2609.40148v1 Announce Type: new Abstract: Power-law learning curves are often treated as fixed properties of a model and its data, although learning-rate and batch-size schedules can change the...
arXiv:2605. 27991v2 Announce Type: replace-cross Abstract: Gradient-flow optimization is usually viewed as an algorithmic procedure for minimizing empirical loss, with training duration selected by validation or heuristic early-stopping rules.
arXiv:2608. 13756v1 Announce Type: new Abstract: Two GPU kernels implementing the same scaled INT8 GEMM interface are usually treated as interchangeable.