arXiv Machine Learning By Nils Matteson

Re-feeding Is Not Replaying: Measuring Replay Noise in Counterfactual Token-Credit Estimation

Read the original on arXiv Machine Learning →

arXiv:2606. 15621v1 Announce Type: new Abstract: Per-token counterfactual credit estimation asks which token in a language-model rollout caused the final answer to be right or wrong: cut the transcript at a pivot, substitute an alternative token, replay continuations, and compare outcomes.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning
Aug 11

CODS: Iterative Bellman-Residual Data Selection for Reusable Offline Reinforcement Learning

arXiv:2608. 07719v1 Announce Type: new Abstract: Offline reinforcement learning repeatedly trains policies from a fixed transition pool, making redundant data costly across seeds and hyperparameters, while naive subsampling can remove rare transitions needed for long-horizon credit assignment.

By Ibne Farabi Shihab, Sanjeda Akter, Abu Sa-Adat Mohamed Moon-Im Al Ahsan, Md Najmus Swaqeeb, Anuj Sharma
arXiv AI
Jul 14

Calibrated e-CUSUM Decoding for Quantized Reasoning Models: Why Token Log-Probability Is the Wrong Observable for Decoding Monitors

arXiv:2607. 11317v1 Announce Type: new Abstract: Low-bit quantization makes small reasoning models inexpensive to deploy but can degrade their chains of thought.

By El Hassane Ettifouri (Novelis Research, Paris, France), Ayoub Belfatmi (Novelis Research, Paris, France), Mahaman Sanoussi Yahaya Alassan (Novelis Research, Paris, France), Walid Dahhane (Novelis Research, Paris, France)