92 papers in our archive carry a September 2026 date, against 74 in May 2026 and an average of 148 per month over the trailing year. (The archive has thin or no coverage for June 2026–August 2026, so those months are excluded from the baseline and the comparison month is the last well-covered one.) This report is generated from the scored archive with no editorial intervention: every number below is a count or an average over our math-complexity and empirical-rigor scores, so it measures what the field submitted, filtered by our quant-relevance gate, not what was eventually published or cited.
Headline numbers
| September 2026 | Trailing year (9 covered months) | |
|---|---|---|
| Papers | 92 | 148 / month |
| Mean empirical rigor | 5.5 / 10 | 5.5 |
| Mean math complexity | 7.3 / 10 | 6.6 |
| Papers with rigor ≥ 8 | 23 (25%) | 23% |
| Papers linking code | 0.0% | 1.7% |
| LLM / NLP share | 8.7% | 11.3% |
Where the papers went
Blue: this month · grey: previous month
Papers can sit in up to three hubs, so shares sum to more than 100%. Measured against the trailing year:
- Gaining share: Options & Derivatives 15% vs 11% (+4.1 pp); Commodities & Energy 5% vs 3% (+2.4 pp); Volatility 17% vs 15% (+2.0 pp)
- Losing share: Risk Management 7% vs 18% (−11.1 pp); Machine Learning 11% vs 20% (−9.5 pp); Market Simulation 2% vs 5% (−2.9 pp)
Theory or evidence?
| Quadrant | September 2026 | Trailing year |
|---|---|---|
| Holy Grail | 51 (55%) | 45% |
| Street Traders | 9 (10%) | 17% |
| Lab Rats | 28 (30%) | 32% |
| Philosophers | 4 (4%) | 5% |
Holy Grail = high math and high rigor; Lab Rats = theory without data; Street Traders = evidence with light theory; Philosophers = neither (definitions). The three most rigorous papers of the month: The Cross-Section of Stock Returns and AI Exposure (rigor 9.0); QuantCode Model: Specializing Language Models for Executable Algorithmic Trading Code (rigor 9.0); Nyström Attention Matches Full Attention for Cross-Sectional Stock Prediction (rigor 8.5).
Top-rated papers of the month
Ranked by rigor-weighted score (60% empirical rigor, 40% math complexity).
- The Cross-Section of Stock Returns and AI Exposure — Holy Grail · Math 7.0 · Rigor 9.0 · NLP & LLMs, HFT & Execution
- Nyström Attention Matches Full Attention for Cross-Sectional Stock Prediction — Holy Grail · Math 7.5 · Rigor 8.5 · Machine Learning, Factor Investing
- Dyson-Schwinger Effective-Action Methods for Rough Volatility: A Correlation-Response Architecture for Calibration, Exotics and Risk — Holy Grail · Math 9.5 · Rigor 7.0 · Volatility, Options & Derivatives
- Target alignment, dilution and forecast selection when cross-sectional forecasts share a common target — Holy Grail · Math 8.5 · Rigor 7.5 · Portfolio Optimization, Factor Investing
- Insights on Time-consistent Deep Hedging under Elicitable Dynamic Risk Measures — Holy Grail · Math 8.5 · Rigor 7.5 · Risk Management, Options & Derivatives
- Jacobian Rank Collapse in Decision-Focused Learning — Holy Grail · Math 7.5 · Rigor 8.0
- Global Multi-Maturity SPX-VIX Calibration Beyond Markovian Stitching — Holy Grail · Math 9.0 · Rigor 7.0 · Volatility
- Hierarchical Multi-Task Learning with Liquidity-Aware Signals for Stock Forecasting — Holy Grail · Math 7.5 · Rigor 8.0 · Machine Learning, Portfolio Optimization
Most active authors
Authors with two or more papers dated this month (author pages exist for researchers with three or more papers in the archive):
- Walter Kurz — 4 papers
- Wojtek Stricker — 2 papers
- Alejandro Rodriguez Dominguez — 2 papers
- Hyungbin Park — 2 papers
- Jun Maeda — 2 papers
Method
Counts cover every paper in the archive whose own date falls in September 2026; late arXiv listings and re-scores can shift a month’s numbers by a few papers after publication. Topic hubs come from a keyword classifier, scores from an LLM-assisted rubric — heuristics, not peer review. The underlying rows are in the downloadable dataset; previous reports are in the reports archive.