# Fallacy Benchmarks Measure Scheme Recognition, Not Fallacy Detection

arXiv: https://arxiv.org/abs/2609.18644 · HTML: https://mlsanity.com/paper/2609.18644/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 27 references that are in the map.
- First posted on arXiv 16 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference](https://mlsanity.com/paper/2412.13663/) (2024)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Probing Neural Network Comprehension of Natural Language Arguments](https://mlsanity.com/paper/1907.07355/) (2019)
- [Argotario: Computational Argumentation Meets Serious Games](https://mlsanity.com/paper/1707.06002/) (2017)
- [CoCoLoFa: A Dataset of News Comments with Common Logical Fallacies Written by LLM-Assisted Crowds](https://mlsanity.com/paper/2410.03457/) (2024)
- [An Explainable Framework for Misinformation Identification via Critical Question Answering](https://mlsanity.com/paper/2503.14626/) (2025)

## Abstract (arXiv)

Fallacy-detection benchmarks pair fallacy classes with a single "valid" or "none" class that takes everything data collection did not label as a fallacy. A detector has two jobs, deciding whether an argument is fallacious and naming which fallacy it commits, and the false-positive rate is meant to measure the first. We show that what these benchmarks actually score is scheme recognition, the ability behind the second job. Their own test sets already show it: when a classifier misses a fallacy, the error lands on "none" rather than on another fallacy type, so detection is failing while classification holds. The reason is what the valid class lacks. The negatives that separate the two jobs are correct arguments using the same argumentation scheme as a fallacy, and they are scarce: nearly absent from the four benchmarks we examined, and rare even under deliberate search. A detector is therefore never tested where recognizing a scheme and judging its use come apart, and can pass on recognition alone. We construct the missing arguments, together with a control condition from the same pipeline that differs only in scheme, so whatever generation contributes, it contributes to both. The classifier labels the scheme-matched negatives as the source fallacy, and labels the wrong-scheme negatives as the scheme they actually use 85.9% of the time and as the source type 0.4%. The classifier has learned which scheme an argument uses, not whether it uses it correctly. The over-flagging follows: a model that scores 16.6% on CoCoLoFa's own valid class flags 58.9% of the constructed arguments. The same dissociation appears in three zero-shot LLM detectors that never saw these benchmarks. We release the items as Scheme Foils. A reported false-positive rate should not be trusted as a measure of detection until the valid class has been audited for scheme-matched coverage.