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.

agents162 research126 infra70 claude69 capital49 labs43 culture39 safety38 openai33 regulation30 anthropic24 security24 reasoning19 evaluation18 robotics13 frontier-models12

arXiv cs.AIPaper9d ago

RollVerify: Bridging Efficiency and Accuracy in Long-Tail Rollout Reinforcement Learning

RL training on LLMs is becoming the production path for reasoning quality, and scaling RL to longer rollouts hits real GPU efficiency walls. RollVerify's repair-before-training approach is more aggressive than reweighting and shows promise for maintaining accuracy while improving throughput. If you're building RL infrastructure, this deserves a test run.

reinforcement-learning llm-training efficiency

arXiv cs.CLPaper11d ago

LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches

This solves a real pain point: RL post-training at scale is memory-bound, and LoGRA makes it practical on hardware most teams have. Enabling a 27B model on a single eight-GPU node is material. For teams already doing RL fine-tuning or considering it, this is a straightforward adoption that frees up budget for longer runs or bigger models. Code is available, implementation is reasonable.

research reinforcement-learning llm-training efficiency

arXiv cs.LGPaper24d ago

PACT: From Credit Assignment to Critic Alignment

This matters if you're training or fine-tuning LLMs with RL. The paper turns credit assignment from intuition into math, showing that response-level REINFORCE and token-level approaches are closer than they appear. For practitioners: it validates the simpler training procedures you're probably already using, and explains why they work. For labs building RL pipelines, this is a foundation to build more efficient training on.

research reinforcement-learning llm-training post-training