AIQF Daily digest — 10 new papers

Quant Paper Radar · 3 of 10

1. 1

   Mixture-of-Experts for Cryptocurrency Order Execution: Training Stability, Tail Risk, and Failure Modes

   Alexander Ardaiz, Varun Budati, Ali Habibnia · ARXIV · Optimal Execution

   quant relevance 0.98

   The paper evaluates reinforcement-learning policies for executing BTC/USDT sell orders, with attention to implementation shortfall, training stability, and tail risk. Its reported results do not establish an execution advantage over TWAP.

   original paper

2. 2

   When a Correct Reward Is Not Enough: Diagnosing and Guiding PPO in an Analytically Solved Broker-Trader Game

   Siu Tung Wong, Carlo Campajola · ARXIV · Optimal execution

   quant relevance 0.91

   The paper tests PPO against an analytical broker execution policy in a simulated broker–trader game. The application concerns execution control, but the results come from a synthetic model rather than measured historical trading performance.

   original paper

3. 3

   Event History Over Scale: Compact Transformers for Low-Latency Limit Order Book Forecasting

   David Schaurecker, Lasse B. Strand, Kevin O'Sullivan et al. · ARXIV · Limit-order-book price forecasting

   quant relevance 0.88

   The paper tests compact models for short-horizon mid-price direction forecasting in equities and intraday electricity markets. It reports classification accuracy and inference latency, not trading returns.

   original paper

7 more papers written up in this period. Open the Quant Paper Radar →