docs: Improved README

Author: lorey

README.md

@@ -1,46 +1,73 @@
-# Apistemic Benchmarks
+# Company-data focused LLM Benchmarks
+We've spent thousands of dollars evaluating LLM performance with company data - so you can skip straight to the results.
+Watch this repo to get a notification as soon as there's a new benchmark or model.
 
 Since we do a lot of LLM-based company analysis at [apistemic](https://apistemic.com),
-we decided to have one central place to keep track of all the benchmarks we do.
+we decided to have one central place to keep track of all the benchmarks.
 This repo thus covers many business/company-related LLM benchmarks.
 
-## Implicit Company Knowledge in Embeddings
-**Goal**: 
-With this benchmark, we want to explore how well LLMs understand companies (and markets).
+## How well do LLMs understand companies?
+Firstly, we want to evaluate how much inherent knowledge LLMs have about companies and markets.
+To do this, we just use company names in all benchmarks without any further context provided.
+
+### Benchmark: Measuring company knowledge inherent in embeddings {#embeddings-benchmark}
+To measure the LLMs' company knowledge in both width and depth,
+we embed company names in this benchmark.
+With the assumption being that the more inherent knowledge an LLM has about companies,
+the more information its embeddings contain.
 
 **Methodology**:
 To measure inherent company knowlege, we prompt the name of companies to get an embedding.
-These embeddings are then used as the only input for a complex regression task, 
-i.e. scoring the competitiveness of two companies.
+These embeddings are then used as the only inputs for a complex regression task, 
+namely scoring the competitiveness of two companies via an SVM.
 A task, that requires a wide and deep understanding of markets, individual companies, business models, and more.
 
 **Dataset**:
-The dataset for this benchmark is provided by [apistemic markets](https://markets.apistemic.com).
-The data contains expert evaluations of competitive positioning between company pairs, 
-where industry professionals assessed relative competitiveness using a standardized five-point scale. 
-These assessments span diverse sectors, encompassing companies of varying sizes and geographic locations 
-to ensure comprehensive coverage across different market contexts.
+See [Competitive Positioning Dataset from Apistemic Markets](#competitive-positioning-dataset).
 
 **Results**:
 ![benchmark of LLM embeddings](.data/plots/r2-scores-boxplot.png)
 
-## LLM-based Company Competitiveness Scoring
-**Goal**: 
-Evaluate how well different LLMs can directly assess company competitiveness compared to human raters.
+### Benchmark: Measuring inherent company knowledge by rating competitiveness {#rating-benchmark}
+As a second benchmark to measure company knowledge, 
+we use the same task as before and prompt the LLMs directly this time.
+We thus provide each LLM with the same instructions a human rater got 
+and ask it to rate the competititveness of two companies.
+Our assumption is that the more knowledge (and understanding) an LLM has, both in width and depth,
+the better it can perform a competitiveness evaluation.
 
 **Methodology**:
-We prompt LLMs to rate the competitiveness between company pairs on a 1-5 scale.
+This benchmark prompts LLMs to rate the competitiveness of company pairs on a 1-5 scale.
 We previously prompted human raters to do the same with the same prompts.
-
 The LLMs receive only company names and must use their internal knowledge to assess competitive relationships. 
-Performance is measured using R² scores and Spearman correlations against expert human evaluations.
-While R² should rate overall similarity to human raters, Spearman correlations between human and LLM ratings should indicate directional correctness,
+Performance is then measured using R² scores and Spearman correlations against expert human evaluations.
+While R² should rate overall similarity to human raters, 
+Spearman correlations between human and LLM ratings should indicate directional correctness,
 i.e. whether the LLM has a sense of competitiveness more generally.
 
 **Dataset**:
-Same expert-evaluated competitive positioning dataset from [apistemic markets](https://markets.apistemic.com).
+See [Competitive Positioning Dataset from Apistemic Markets](#competitive-positioning-dataset).
 
 **Results**:
 ![LLM R² scores](.data/plots/r2-scores-barplot.png)
 
 ![LLM Spearman correlations](.data/plots/spearman-correlations-barplot.png)
+
+## Datasets
+
+Our benchmarks are based on proprietary datasets.
+This sections covers a description of each dataset used.
+
+### Competitive Positioning Dataset from Apistemic Markets {#competitive-positioning-dataset}
+
+**Source**: [apistemic markets](https://markets.apistemic.com)
+
+**Description**: 
+Expert evaluations of competitive positioning between company pairs, 
+where industry professionals assessed relative competitiveness using a standardized five-point scale. 
+These assessments span diverse sectors, encompassing companies of varying sizes and geographic locations 
+to ensure comprehensive coverage across different market contexts.
+
+**Used in**:
+- [Benchmark: Measuring company knowledge inherent in embeddings](#embeddings-benchmark)
+- [Benchmark: Measuring inherent company knowledge by rating competitiveness](#rating-benchmark)