Jesse A. Rodríguez | Searcharxiv

arXiv2026

LaTA: A Drop-in, FERPA-Compliant Local-LLM Autograder for Upper-Division STEM Coursework

Large-language-model (LLM) graders promise to relieve the grading burden of upper-division STEM courses, but most deployments to date send student work to third-party APIs, violating FERPA and exposing institutions to data risk while requiring substantial assignment modification. We present $\textbf{LaTA}\ (\textit{LaTeX Teaching Assistant})$, a drop-in, open-source autograder that runs entirely on commodity on-premises hardware and assumes a LaTeX-native workflow already adopted by many engineering and physics courses. LaTA implements a four-stage pipeline (ingest, segment, grade, report) using a locally hosted open-weight chain-of-thought LLM grader (gpt-oss:120b) that compares student work to an instructor-authored reference solution and applies a YAML rubric with binary per-item scoring. We deployed LaTA in Winter~2026 in ME 373 (Mechanical Engineering Methods) at Oregon State University, grading every weekly assignment for approximately 200 students on a single Mac Studio at \$0 marginal cost per assignment and 1--3 minutes of wall-clock time per submission, enabling regrading of corrected assignments and greatly expanded TA office hour offerings. The instructor-confirmed grading-error rate held at roughly $0.02$--$0.04\%$ per rubric line item across the term. Relative to the same instructor's previous traditionally-graded cohort, the LaTA-graded cohort outperformed by approximately $11\%$ on the midterm exam and $8\%$ on the final exam, and reported large gains in self-assessed confidence on every stated learning objective ($N = 159$ survey responses, $Δ\geq +1.49$ Likert points, $p < 10^{-27}$ on every comparison). We release the code under AGPLv3.

cs.AI↗

arXiv2026

In-Situ Inverse Design of a Plasma Metamaterial Beam Steering Device

Inverse design is a commonly used methodology for creating devices that manipulate electromagnetic (EM) waves by algorithmically modifying device parameters to achieve a desired functionality. Utilizing plasma, a dynamically tunable medium, allows the optimization of the design process to be conducted directly on the experimental hardware (in-situ). A key advantage of this method is the creation of devices that are inherently switchable and dynamically reconfigurable. Bayesian optimization is used to tune the plasma density of 91 independent discharges that make up a plasma metamaterial (PMM) device to steer incoming EM waves to desired exit waveguides. Measurements were conducted in an automated loop where a vector network analyzer records the PMM transmission characteristics for each device setting. By relying only on measured scattering parameters, this gradient-free approach is robust to experimental drift and noise and does not require complex full-wave models. Significant performance improvements over traditional simulation-based (in-silico) inverse design are demonstrated, with in-situ Bayesian optimization achieving up to 10,000x higher isolation between ports than the best in-silico design at the same target frequency. This work also presents guidelines for applying Bayesian optimization to noisy, high-dimensional physical systems.

physics.optics↗