Eckhardt, Paradoxes in Probability Theory.

One handy quote:

Newcomb’s problem has fractured decision theory into a host of warring parties and engendered formulations that are complicated, inelegant, and, I would venture, incorrect. These newer theories share a common trait: they are all more or less self-conscious attempts to secure a two-boxer resolution to Newcomb’s problem. Most disagreement between experts on this subject concerns the correct way to reach this conclusion. This entire edifice, its concordances and its disputes, are vulnerable to the possibility that two-boxing is the wrong way to play.

Eckhardt proposes a new decision theory, "coherent decision theory" (which annoyingly has the same acronym as causal decision theory).