Research

Evaluate Tent recovery artifacts for numeric metrics, source-boundary compliance, and mechanism-faithful proxy evidence.

Run a bounded CMA-ES proxy experiment that validates sampling, selection, CSA, and covariance adaptation evidence.

Run a deterministic reduced Gaussian-mixture recovery that exercises stochastic-interpolant construction, objectives, denoising, and sampling evidence.

Construct endpoint-valid stochastic interpolant samples with latent schedules and derivative targets for reduced or full generative recovery experiments.

Evaluate SR3-style proxy runs with consistency metrics, source-boundary records, and mechanism checks. This reusable skill supports bounded SR3 recovery experiments.

Build and validate SR3-style low-resolution conditioning paired with high-resolution targets for reduced or full recovery experiments.

Define continuous-time score-SDE schedules, perturbation kernels, reverse drifts, and probability-flow checks for recovery experiments.

Run a bounded synthetic MSFM proxy experiment that invokes generated coupling and Joint CFM loss skills.

Evaluate whether an MSFM recovery result exercised BatchOT coupling, Joint CFM loss, optimizer evidence, and soft-mode proxy constraints.

Run a bounded DDPM proxy experiment that exercises generated schedule, epsilon-loss, and reverse-step skills.

Run and validate a bounded GeDi proxy recovery that invokes generated GeDi modules and records mechanism-faithful evidence.

Evaluate coherence boosting recovery metrics and source-boundary mechanism checks for proxy experiments.

Run a bounded mechanism-faithful TSNPE proxy recovery experiment.

Compute a deterministic classifier two-sample accuracy proxy for comparing posterior sample distributions.

Run an auditable reduced recovery experiment for the SBI benchmark by composing generated task, sampling, and C2ST skills.

Execute a bounded soft-mode proxy experiment that validates P3O mechanisms with generated skills and numeric training evidence.

Validate mechanism evidence for adversarial reprogramming proxy or full recovery experiments.

Build and validate harmless corpus splits that mirror visual jailbreak optimization and held-out evaluation protocols.

Compute held-out benign-versus-adversarial obedience deltas and mechanism checks for safe jailbreak proxies.

Run an executable reduced TeCoA recovery experiment with generated skill invocations and mechanism checks.

Build and validate OpenCLIP-style scale tables for CLIP scaling-law experiments, including total compute and error metric derivation.

Build LLaVA-style visual instruction records from captions and boxes for mechanism-faithful recovery experiments.

Run a bounded soft-mode CLIP proxy recovery that logs contrastive training and zero-shot evaluation evidence.

Specify and validate the robust min-max objective used for PGD adversarial training from Madry et al. 2017.