Category: Universal Intelligence Benchmark

Inference-agnostic intelligence measurement framework. Meta-research and novel benchmarks for AI beyond text generation.

Causal Intelligence as a UIB Dimension: Measuring What Models Actually Understand

Posted on by
Benchmark Research
Benchmark Research by Oleh Ivchenko  ·  DOI: 10.5281/zenodo.19102383  71stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources7%○≥80% from editorially reviewed sources
[t]Trusted73%○≥80% from verified, high-quality sources
[a]DOI53%○≥80% have a Digital Object Identifier
[b]CrossRef7%○≥80% indexed in CrossRef
[i]Indexed53%○≥80% have metadata indexed
[l]Academic80%✓≥80% from journals/conferences/preprints
[f]Free Access93%✓≥80% are freely accessible
[r]References15 refs✓Minimum 10 references required
[w]Words [REQ]1,867✓Minimum 1,000 words of article body. Current: 1,867
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]9✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 9
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19102383
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness27%○≥60% of references from 2025–2026. Current: 27%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (63 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)

Current AI benchmarks predominantly measure pattern recognition and statistical correlation — capabilities that, while impressive, fall short of genuine understanding. This article introduces Causal Intelligence as a formal dimension within the Universal Intelligence Benchmark (UIB) framework, arguing that any credible measure of machine intelligence must evaluate whether systems can reason abo...

Show moreHide
Benchmark Research by Oleh Ivchenko DOI: 10.5281/zenodo.19102383 71stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources7%○≥80% from editorially reviewed sources
[t]Trusted73%○≥80% from verified, high-quality sources
[a]DOI53%○≥80% have a Digital Object Identifier
[b]CrossRef7%○≥80% indexed in CrossRef
[i]Indexed53%○≥80% have metadata indexed
[l]Academic80%✓≥80% from journals/conferences/preprints
[f]Free Access93%✓≥80% are freely accessible
[r]References15 refs✓Minimum 10 references required
[w]Words [REQ]1,867✓Minimum 1,000 words of article body. Current: 1,867
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]9✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 9
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19102383
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness27%○≥60% of references from 2025–2026. Current: 27%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (63 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)
Universal Intellig…Read More
Read more

Inference-Agnostic Intelligence: The UIB Theoretical Framework

Posted on by
Benchmark Research
Benchmark Research by Oleh Ivchenko  ·  DOI: 10.5281/zenodo.19064304  70stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources11%○≥80% from editorially reviewed sources
[t]Trusted74%○≥80% from verified, high-quality sources
[a]DOI42%○≥80% have a Digital Object Identifier
[b]CrossRef5%○≥80% indexed in CrossRef
[i]Indexed68%○≥80% have metadata indexed
[l]Academic63%○≥80% from journals/conferences/preprints
[f]Free Access74%○≥80% are freely accessible
[r]References19 refs✓Minimum 10 references required
[w]Words [REQ]2,121✓Minimum 1,000 words of article body. Current: 2,121
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]9✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 9
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19064304
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness33%○≥60% of references from 2025–2026. Current: 33%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (62 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)

Current AI benchmarks measure narrow task performance — accuracy on question sets, code generation pass rates, or image recognition scores. They rarely ask the deeper question: what is intelligence, and how should we measure it independent of the hardware, API, or inference provider running the model? This article proposes the Universal Intelligence Benchmark (UIB) theoretical framework: an eig...

Show moreHide
Benchmark Research by Oleh Ivchenko DOI: 10.5281/zenodo.19064304 70stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources11%○≥80% from editorially reviewed sources
[t]Trusted74%○≥80% from verified, high-quality sources
[a]DOI42%○≥80% have a Digital Object Identifier
[b]CrossRef5%○≥80% indexed in CrossRef
[i]Indexed68%○≥80% have metadata indexed
[l]Academic63%○≥80% from journals/conferences/preprints
[f]Free Access74%○≥80% are freely accessible
[r]References19 refs✓Minimum 10 references required
[w]Words [REQ]2,121✓Minimum 1,000 words of article body. Current: 2,121
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]9✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 9
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19064304
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness33%○≥60% of references from 2025–2026. Current: 33%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (62 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)
Universal Intellig…Read More
Read more

The Measurement Crisis: Saturation, Goodhart’s Law, and the End of AI Leaderboards

Posted on by
Benchmark Research
Benchmark Research by Oleh Ivchenko  ·  DOI: 10.5281/zenodo.19007432  56stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources0%○≥80% from editorially reviewed sources
[t]Trusted40%○≥80% from verified, high-quality sources
[a]DOI40%○≥80% have a Digital Object Identifier
[b]CrossRef0%○≥80% indexed in CrossRef
[i]Indexed40%○≥80% have metadata indexed
[l]Academic40%○≥80% from journals/conferences/preprints
[f]Free Access80%✓≥80% are freely accessible
[r]References5 refs○Minimum 10 references required
[w]Words [REQ]3,347✓Minimum 1,000 words of article body. Current: 3,347
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]1✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 1
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19007432
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness33%○≥60% of references from 2025–2026. Current: 33%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (39 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)

The AI evaluation ecosystem is in crisis. Frontier models now exceed 90% accuracy on MMLU, 95% on HumanEval, and 93% on HellaSwag — scores that were considered unattainable three years ago. This saturation is not evidence of intelligence; it is evidence that our instruments have failed. We argue that three convergent forces have rendered current AI leaderboards meaningless: (1) benchmark satura...

Show moreHide
Benchmark Research by Oleh Ivchenko DOI: 10.5281/zenodo.19007432 56stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources0%○≥80% from editorially reviewed sources
[t]Trusted40%○≥80% from verified, high-quality sources
[a]DOI40%○≥80% have a Digital Object Identifier
[b]CrossRef0%○≥80% indexed in CrossRef
[i]Indexed40%○≥80% have metadata indexed
[l]Academic40%○≥80% from journals/conferences/preprints
[f]Free Access80%✓≥80% are freely accessible
[r]References5 refs○Minimum 10 references required
[w]Words [REQ]3,347✓Minimum 1,000 words of article body. Current: 3,347
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]1✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 1
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19007432
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness33%○≥60% of references from 2025–2026. Current: 33%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (39 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)
Universal Intellig…Read More
Read more

The Meta-Meta-Analysis: A Systematic Map of What 200 AI Benchmark Studies Actually Measured

Posted on by
Benchmark Research
Benchmark Research by Oleh Ivchenko  ·  DOI: 10.5281/zenodo.19001033  53stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources6%○≥80% from editorially reviewed sources
[t]Trusted39%○≥80% from verified, high-quality sources
[a]DOI11%○≥80% have a Digital Object Identifier
[b]CrossRef0%○≥80% indexed in CrossRef
[i]Indexed39%○≥80% have metadata indexed
[l]Academic39%○≥80% from journals/conferences/preprints
[f]Free Access44%○≥80% are freely accessible
[r]References18 refs✓Minimum 10 references required
[w]Words [REQ]2,576✓Minimum 1,000 words of article body. Current: 2,576
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]14✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 14
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19001033
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness32%○≥60% of references from 2025–2026. Current: 32%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (33 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)

We present a meta-meta-analysis of 217 benchmark evaluation studies published between 2020 and 2026, examining not the benchmarks themselves but the systematic reviews that assess them. Our coverage matrix reveals a profound structural bias: 78.3% of surveyed studies evaluate text-based capabilities, while causal reasoning (4.1%), embodied intelligence (1.8%), and social cognition (0.9%) remain...

Show moreHide
Benchmark Research by Oleh Ivchenko DOI: 10.5281/zenodo.19001033 53stabilfr·wuedophcgmx
BadgeMetricValueStatusDescription
[s]Reviewed Sources6%○≥80% from editorially reviewed sources
[t]Trusted39%○≥80% from verified, high-quality sources
[a]DOI11%○≥80% have a Digital Object Identifier
[b]CrossRef0%○≥80% indexed in CrossRef
[i]Indexed39%○≥80% have metadata indexed
[l]Academic39%○≥80% from journals/conferences/preprints
[f]Free Access44%○≥80% are freely accessible
[r]References18 refs✓Minimum 10 references required
[w]Words [REQ]2,576✓Minimum 1,000 words of article body. Current: 2,576
[u]Original [REQ]100%✓≥85% unique sentences (no padding or repeated blocks) and no leaked drafting notes. Current: 100% unique
[e]Evidence [REQ]14✓Cites at least 1 external source(s): links, DOIs or a reference list. Current: 14
[d]DOI [REQ]✓✓Zenodo DOI registered for persistent citation. DOI: 10.5281/zenodo.19001033
[o]ORCID✓✓Author ORCID verified for academic identity
[p]Peer Reviewed—○Peer reviewed by an assigned reviewer
[h]Freshness32%○≥60% of references from 2025–2026. Current: 32%
[c]Data Charts0○Original data charts from reproducible analysis (min 2). Current: 0
[g]Code—○Source code available on GitHub
[m]Diagrams3✓Mermaid architecture/flow diagrams. Current: 3
[x]Cited by0○Referenced by 0 other hub article(s)
Score = Ref Trust (33 × 60%) + Required (4/4 × 30%) + Optional (2/7 × 10%)
Universal Intellig…Read More
Read more