To KL Diverge, or Not to KL Diverge: A Question for Quants
Summary
Discusses the trade-offs of using Kullback-Leibler divergence in quantitative analysis, framing it as a Hamlet-like dilemma for quants.
Similar Articles
KL Zero: KL divergence intuition game
KL Zero is an interactive browser game where players draw a probability distribution to match a target KL divergence value, helping users intuitively understand the concept of KL divergence in machine learning.
ByteShape Qwen 3.8 27B: To KL Diverge or Not to KL Diverge, Part 2: Metric Boogaloo
ByteShape released ShapeLearn GGUFs for Qwen 3.8 27B, achieving high accuracy on benchmarks, and discussed the role of KLD in quantization, with a related paper accepted at EMNLP 2026.
How Wrong Can a Good Predictor Be? Diverging Updates with Vanishing Predictive KL
This paper proves that unbounded gaps between update maps can coexist with vanishing predictive KL divergence in stationary symmetric Gaussian HMMs, showing that internal computational mismatches do not necessarily imply predictive failure.
Statistical Inference and Quality Measures of KV Cache Quantisations Inspired by TurboQuant
This paper analyzes KV cache quantization schemes inspired by TurboQuant, using statistical inference and a new 6D error framework to evaluate quality measures like KL divergence and geometric error.
Not Every Divergence Should Be Suppressed: Counterfactual Recoverability in On-Policy Distillation
Presents counterfactual recoverability as an outcome-grounded decision variable for on-policy distillation, showing that selectively supervising based on recoverability outperforms divergence-only methods across AIME and GPQA benchmarks.