FLD (Formal Logic Deduction)

Hitachi's synthetic deduction set: given invented facts and a hypothesis, choose proved, disproved, or unknown without using world knowledge.

Also known as: Formal Logic Deduction, FLD.v2, FLD-star, FLD★

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categoryreasoning
Subcategorysynthetic multi-step formal-logic deduction (answer classification)
Page statusunknown
Metricexact_match
Directionhigher_is_better
Unit%
Dataset size5000
Dataset licenceCC-BY-4.0
PublisherHitachi, Ltd. (hitachi-nlp)

What it measures

FLD (Formal Logic Deduction) asks a model to (dis)prove a hypothesis from a set of facts that are logically structured but semantically invented, so memorised world knowledge should not help. The ICML 2023 paper scores both a proof trace and a three-way answer (proved / disproved / unknown). EleutherAI lm-eval ships the simplified "answer accuracy" setting only: generate the world_assump_label (Hub gold is uppercase PROVED, DISPROVED, or UNKNOWN). English natural-language facts, plus parallel logical-formula prompts. FLD★ (star / FLD.4) uses deeper trees (up to depth 8 vs up to 3) and is harder. Not [logical_deduction](logical_deduction.md) and not [formal_fallacies_syllogisms_negation](formal_fallacies_syllogisms_negation.md).

Task format

lm-eval default prompt: "Based on the provided facts ($context$), either prove or disprove the hypothesis or state that it is unknown." plus prompt_serial. Formula tasks substitute context_formula and hypothesis_formula. Target is world_assump_label. Metric is exact_match after stripping whitespace and taking the first line.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub