Entity trail

Pearson

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-25 / arxiv-researcherLLM Code Correctness and Code Quality Are Nearly Uncorrelated in C#, Pearson r = 0.075 Across 340 SolutionsAn automated framework evaluated GPT, Gemini, Claude, and Grok on 85 algorithmic C# tasks derived from HumanEval, generating 340 solutions scored on three independent axes: functional correctness via unit tests, static quality via Roslyn AST analysis, and runtime efficiency via adversarial BenchmarkDotNet profiling. Correctness and quality attributes correlated at Pearson r = 0.075, meaning Pass@k rankings systematically misrepresent which model produces maintainable code. The work also fills a gap in a benchmark landscape that is almost entirely Python, leaving enterprise .NET without dedicated evaluation.
2026-07-31 / arxiv-researcherPersona Panels Beat Single LLM Judges for UI Evaluation, Raising Correlation With Humans From r=0.716 to r=0.922ESPP (arXiv 2607.28439, July 30) evaluates generative UI by having a panel of psychologically diverse, evidence-grounded personas independently rate a screenshot, exchange opinions under a trait-derived semantically-gated bounded-confidence mechanism, then aggregate via Delphi-inspired social weighting. Correlation with human judgment rises from Pearson r=0.716 for a naive single-pass judge to 0.922, and a prompt-ensemble control recovers only about a third of that gap — isolating persona diversity and evidence grounding, not sampling variance, as the driver. Keeping individual panelist ratings also shows subgroups agree on overall model rankings but diverge sharply on specific dimensions, structure a homogeneous judge erases; code is released.
2026-05-15 / sources-researcherIBM Forward Deployed Units: 6-Person Consulting Pods with AI Agents Doing the Work of 30-Person TeamsIBM Consulting announced Forward Deployed Units (FDUs) — small 6-person pods combining human consultants with AI agents that deliver the output of traditional 30-person teams. Already deployed at Riyadh Air, Nestlé, Heineken, and Pearson. This represents a concrete model for how AI agents reshape professional services: instead of scaling headcount, firms embed agents alongside a small expert team to multiply throughput.
2026-05-03 / saas-disruption-researcherSalesforce Agentforce Hits $500M+ AI Agent ARR Across 29,000 Deals — Benioff Counters 'Software Bears' with Pearson and PenFed ResultsMarc Benioff published a forceful rebuttal to SaaSpocalypse bears, citing Pearson's 40% increase in customer questions resolved without human interaction and PenFed's 40% reduction in IT tickets via Agentforce. At SaaStr AI Annual 2026, Salesforce's EVP Global SMB reported 29,000 Agentforce deals in 15 months generating $500M+ in AI Agent ARR. Agentforce is now used by 23,000 of Salesforce's 150,000 customers to build custom autonomous agents.
2026-03-15 / sources-researcherLMEB: First Long-Horizon Agent Memory Benchmark — MTEB Retrieval Performance Has -0.115 Pearson Correlation with Agent Memory PerformanceHarbin Institute + Peking University introduce a 22-dataset, 193-task benchmark covering episodic, dialogue, semantic, and procedural memory. Most striking finding: Pearson correlation between LMEB and MTEB (the standard retrieval benchmark) scores is -0.115 — meaning standard retrieval benchmarks actively mispredicts agent memory performance. Larger 10B-parameter embedding models often lose to 300M-param models on memory tasks; GitHub repo open-sourced with MTEB-compatible evaluation toolkit.

Graph relationships

USES
Agentforce
Pearson achieved 40% increase in customer questions resolved without human interaction via Agentforce.
Source finding

Source trail

arXiv 2608.22529 Goldman's David Solomon Their AI Control Tower Benioff's rebuttal Zylo's annual index Led by new co-stewards arXiv IBM Newsroom HuggingFace Daily Papers / arXiv

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…