An on-premises LLM autograder graded a 200-student course at 1–3 minutes wall-clock time per submission across a full term¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Across eight homework assignments, LaTA graded the 200-student roster in 1–3 minutes of wall-clock time per submission, with per-assignment totals driven by problem count and solution length rather than class size. → Jesse A. Rodríguez 2026
Evidence¶
Jesse A. Rodríguez 2026¶
Jesse A. Rodríguez. (2026). LaTA: A Drop-In, FERPA-Compliant Local-LLM Autograder for Upper-Division STEM Coursework. arXiv preprint. https://arxiv.org/abs/2605.05410 doi:10.48550/arxiv.2605.05410
q2 · i? · design · r2
Operational evidence collected by the instructor during the Winter 2026 term in ME 373, recorded as the standard output of each grading run. The quote reports the printed per-submission wall-clock time on the single Mac Studio host.
"LaTA graded the 200-student roster in 1–3 minutes of wall-clock time per submission on the Mac Studio M3 Ultra described in §4.1, with the per-assignment total dominated by the number of problems and the length of the worked solutions rather than by class size."
Discussion¶
Related Claims¶
- The between-cohort exam gain is attributable only to a composite intervention (autograding plus corrections plus expanded office hours), not to LaTA alone — related
- Instructor-confirmed grading errors occurred at roughly 0.02–0.04% of rubric line items across the term — related
- Student perceptions of the autograder were mixed, with the LaTeX workflow rated more positively than the grader itself — related
- Students reported first-pass false-positive deductions at a mean of 2.94 per term, normalising to about 0.6% of rubric items — related
- Student-reported LaTeX time overhead fell about 46% across the term, from 87 to 47 minutes per assignment — related
- A page-by-page, rubric-guided multimodal LLM workflow graded an authentic handwritten exam in about three hours at roughly $100 in token costs, versus about $3,500 in TA time — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
lata-onprem-autograder-throughput |
title |
An on-premises LLM autograder graded a 200-student course at 1–3 minutes wall-clock time per submission across a full term |
description |
An on-premises LLM autograder graded a 200-student course at 1–3 minutes wall-clock time per submission across a full term |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — jesse-a-rodríguez-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.