The Signal

Everything that matters in AI, with our take.

Updated through the day. Every headline links straight to the source. The two lines underneath are ours.

research1156 agents716 infra561 labs317 culture268 capital253 regulation233 claude175 openai149 safety103 security89 evaluation77 reasoning61 alignment52 benchmarks50 interpretability48

arXiv cs.LGPaper13d ago

Subspace Inference Enables Efficient Active Reward Learning from Preferences

The core idea—active learning needs uncertainty quantification, and full posterior over large networks is expensive—is sound, but the execution is specialist. If you're training reward models at scale with human feedback, this Kalman filtering approach could cut query counts. For most teams, vanilla RLHF still works; this is an optimization for high-volume settings.

research rlhf active-learning reward-modeling

arXiv cs.LGPaper13d ago

DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training

This tackles a real problem: rubric-based rewards are coarse, but you need fine-grained signal for step-level learning. The closed-form redistribution is elegant and the results on AppWorld suggest it matters. If you're training agents with outcome-blind rewards, this is worth implementing, though the benchmark is still narrow.

agents reinforcement-learning reward-modeling long-horizon