AIQF Weekly digest — 1 strategy, 69 new papers

The week in numbers

Max drawdown -2.4%, -7.0% at market β · 21 strategies live · as of 2026-10-16

At market β restates the record at the market's beta — the same strategies held at 3.70× the exposure: the return is multiplied by that factor, the drawdown re-measured on the scaled line, and Sharpe is unchanged by it. Arithmetic on the record, not a second track record — no leverage was used and no financing cost is modelled.

New public strategies

• SPY-Relative Drawdown-Targeted Constrained Kelly Overlay

  Daily estimates a constrained Kelly allocation across SPY and GLD, then scales its active deviation from a 100% SPY benchmark to prescribed drawdown-risk targets. Infeasible rebalances are skipped rather than clipped or partially executed.

  Sharpe 0.67 · total return 59.2% · max drawdown -35.9%

Quant Paper Radar · 5 of 69

1. 1

   From Economic Explanation to Performance Improvement in Machine-Learning Portfolios

   Yunqi Liang, Stefan Zohren, Mihai Cucuringu · ARXIV · Machine-learning statistical arbitrage and portfolio attribution

   we backtested it — Sharpe -0.82 · quant relevance 1.00

   The paper uses portfolio-performance attributions to adjust machine-learning statistical-arbitrage allocations without retraining. It reports improved out-of-sample net Sharpe ratios for US equity portfolios, but our shorter data history cannot reproduce its full 2006–2024 test.

   original paper

2. 2

   AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery

   Jie Huang, Yuecheng Hong, Yao Chen et al. · ARXIV · Symbolic factor discovery

   we backtested it — Sharpe 0.83 · quant relevance 0.98

   AlphaRLM trains a language model to generate symbolic stock-selection factors and reports improved backtested excess returns. Its US-equity factor evaluation is testable here, but the literature-conditioned reinforcement-learning method and Chinese-market results are not reproducible with the stated resources.

   original paper

3. 3

   From Alpha to Execution: Trading Horizons, Regimes, and Schedules under Concave Impact and Hurst Scaling

   Dnyanesh Kulkarni, Revant Nayar · ARXIV · Optimal Execution

   we backtested it — Sharpe -0.29 · quant relevance 0.98

   The paper derives an alpha-aware execution horizon and a front-loaded schedule under concave impact and Hurst-scaled price risk. Its reported 10–17% cost saving comes from the execution-cost model, not measured trading performance.

   original paper

4. 4

   Reading the News Without Hindsight: NoLBERT Topics and the Cross-Section of Stock Returns

   Kevin Foley, Jonathan Hartadi, Shivesh Prakash et al. · ARXIV · News-based cross-sectional return prediction

   we backtested it — Sharpe -0.73 · quant relevance 0.98

   The paper tests news-topic attention and stock exposures as cross-sectional return signals for US equities. Its reported ensemble performance does not significantly beat a price-only benchmark. A forward-shifted version is backtestable, but the platform cannot reproduce the early training sample or the NoLBERT comparison.

   original paper

5. 5

   SBCA: cross-modal gated actor-critic with frozen BERT-based sentiment features for multi-asset portfolio optimization

   Jinfeng Pan, Mi Liu, 晓萍 et al. · OPENALEX · News-informed portfolio optimization

   we backtested it — Sharpe 2.62 · quant relevance 0.98

   The paper tests a daily, news-informed actor-critic policy for allocating among U.S. equities, with downside-risk and turnover penalties. Its reported performance applies to its own small equity portfolios; our shorter news history cannot reproduce its 11-year test.

   original paper

69 new papers written up in this period. Open the Quant Paper Radar →