MétaCan
Menu
Back to cohort

Hence, Socrates is mortal: A Benchmark for Natural Language Syllogistic Reasoning

2023· article· en· W4385572407 on OpenAlexaff
Yongkang Wu, Meng Han, Yutao Zhu, Lei Li, Xinyu Zhang, Ruofei Lai, Xiaoguang Li, Yuanhang Ren, Zhicheng Dou, Zhao Cao

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsUniversité de Montréal
Fundersnot available
KeywordsSyllogismComputer scienceNatural language processingArtificial intelligenceParaphraseNatural language understandingNatural languageConstruct (python library)Deductive reasoningBenchmark (surveying)Programming languageLinguistics

Abstract

fetched live from OpenAlex

Syllogistic reasoning, a typical form of deductive reasoning, is a critical capability widely required in natural language understanding tasks, such as text entailment and question answering.To better facilitate research on syllogistic reasoning, we develop a benchmark called SYLLOBASE that differs from existing syllogistic datasets in three aspects: (1) Covering a complete taxonomy of syllogism reasoning patterns; (2) Containing both automatically and manually constructed samples; and (3) Involving both the generation and understanding tasks.We automatically construct 50k template-based syllogism samples by mining syllogism patterns from Wikidata and ConceptNet.To improve our dataset's naturalness and challenge, we apply GPT-3 to paraphrase the templatebased data and further manually rewrite 1,000 samples as the test set.State-of-the-art pretrained language models can achieve the best generation ROUGE-L of 38.72 by T5 and the best multi-choice accuracy of 72.77% by RoBERTa on SYLLOBASE, which indicates the great challenge of learning diverse syllogistic reasoning types on SYLLOBASE.Our datasets are released at https://github.com/ casually-PYlearner/SYLLOBASE.ConceptNet An open, multilingual knowledge graph (1) Triplet Extraction (Section 3.1) (2) Syllogism Construction (Section 3.2) (3) Paraphrasing (GPT-3) (Section 3.3) ① (human, capable of, mortal), (Socrates, is a, human) ② (meteoritics, subclass of, astronomy), (astronomy, subclass of, exact science) Premise 1: All human are mortal.Premise 1: Some astronomy are meteoritics.Premise 2: All Socrates are human.Premise 2: All astronomy are exact science.Conclusion: All Socrates are mortal.Conclusion: Some exact science are meteoritics Pattern: All 𝑚 are 𝑝, all 𝑠 are m ⟶ All 𝑠 are 𝑝.Pattern: Some 𝑚 are 𝑝, all 𝑚 are 𝑠 ⟶ Some 𝑠 are 𝑝.Premise 1: It is a fact that all human beings are mortal.Premise 2: Socrates was a classic Greek philosopher credited as one of the founders of Western philosophy.Conclusion: It is true that Socrates was mortal as well.Premise 1: Meteoritics, which is a specific type of astronomy, involves the study of meteors and meteorites.Premise 2: Astronomy is an exact science meaning precise observation and mathematical calculations are implemented.Conclusion: Some precise sciences are used to further expand upon knowledge of meteorites and meteors.Table 2: Examples of syllogisms from our test set.Categorical Syllogism Premise 1: Carbon dioxide is a chemical compound.Premise 2: Chemical compounds are considered pure substances.Conclusion: Pure substances include carbon dioxide.Hypothetical Syllogism Premise 1: When you make progress in your project, you may want to celebrate.Premise 2: Having a party is a good choice if you want to celebrate.Conclusion: You may want to have a party if you achieve great progress in your project.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.022
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.028
Threshold uncertainty score0.055

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.022
Meta-epidemiology (narrow)0.0040.001
Meta-epidemiology (broad)0.0010.003
Bibliometrics0.0080.007
Science and technology studies0.0020.001
Scholarly communication0.0040.007
Open science0.0060.003
Research integrity0.0030.004
Insufficient payload (model declined to judge)0.0150.017

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.021
GPT teacher head0.294
Teacher spread0.273 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2023
Admission routes1
Has abstractyes

Explore more

Same topicTopic ModelingFrench-language works237,207