Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Autoscoring Anticlimax: A Meta-analytic Understanding of AI's Short-answer Shortcomings and Wording Weaknesses

Michael Hardy

Published

Mar 5, 2026

Citations

0

Trust level

Low

Usefulness score

0/100 (Low)

Extraction confidence

35% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Domain Experts

Signals refreshed

Mar 26, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this as background context only. Do not make protocol decisions from this page alone.

Best use

Background context only

Use if you need

A secondary eval reference to pair with stronger protocol papers.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

This paper looks adjacent to evaluation work, but not like a strong protocol reference.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
0/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Automated short-answer scoring lags other LLM applications. We meta-analyze 890 culminating results across a systematic review of LLM short-answer scoring studies, modeling the traditional effect size of Quadratic Weighted Kappa (QWK) with mixed effects metaregression. We quantitatively illustrate that that the level of difficulty for human experts to perform the task of scoring written work of children has no observed statistical effect on LLM performance. Particularly, we show that some scoring tasks measured as the easiest by human scorers were the hardest for LLMs. Whether by poor implementation by thoughtful researchers or patterns traceable to autoregressive training, on average decoder-only architectures underperform encoders by 0.37--a substantial difference in agreement with humans. Additionally, we measure the contributions of various aspects of LLM technology on successful scoring such as tokenizer vocabulary size, which exhibits diminishing returns--potentially due to undertrained tokens. Findings argue for systems design which better anticipates known statistical shortcomings of autoregressive models. Finally, we provide additional experiments to illustrate wording and tokenization sensitivity and bias elicitation in high-stakes education contexts, where LLMs demonstrate racial discrimination. Code and data for this study are available.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

missing

None explicit

No explicit feedback protocol extracted.

"Automated short-answer scoring lags other LLM applications."

Evaluation Modes

partial

Automatic Metrics

Includes extracted eval setup.

"Automated short-answer scoring lags other LLM applications."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Automated short-answer scoring lags other LLM applications."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"Automated short-answer scoring lags other LLM applications."

Reported Metrics

partial

Kappa

Useful for evaluation criteria comparison.

"We meta-analyze 890 culminating results across a systematic review of LLM short-answer scoring studies, modeling the traditional effect size of Quadratic Weighted Kappa (QWK) with mixed effects metaregression."

Rater Population

partial

Domain Experts

Helpful for staffing comparability.

"We quantitatively illustrate that that the level of difficulty for human experts to perform the task of scoring written work of children has no observed statistical effect on LLM performance."

Benchmarks and datasets

No benchmark or dataset names were extracted from the available abstract.

Reported metrics

kappa
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Domain Experts
Expertise required
Coding
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

Automated short-answer scoring lags other LLM applications.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Automated short-answer scoring lags other LLM applications.
  • We meta-analyze 890 culminating results across a systematic review of LLM short-answer scoring studies, modeling the traditional effect size of Quadratic Weighted Kappa (QWK) with mixed effects metaregression.
  • We quantitatively illustrate that that the level of difficulty for human experts to perform the task of scoring written work of children has no observed statistical effect on LLM performance.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Check the full text for explicit evaluation design choices (raters, protocol, and metrics).
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Agreement / Kappa

Contribution summary

  • We quantitatively illustrate that that the level of difficulty for human experts to perform the task of scoring written work of children has no observed statistical effect on LLM performance.
  • Particularly, we show that some scoring tasks measured as the easiest by human scorers were the hardest for LLMs.
  • Whether by poor implementation by thoughtful researchers or patterns traceable to autoregressive training, on average decoder-only architectures underperform encoders by 0.37--a substantial difference in agreement with humans.

Why it matters for eval

  • We quantitatively illustrate that that the level of difficulty for human experts to perform the task of scoring written work of children has no observed statistical effect on LLM performance.
  • Particularly, we show that some scoring tasks measured as the easiest by human scorers were the hardest for LLMs.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Metric reporting is present

    Detected: kappa