Post-Training · AI Beat

14 Aug 2026 · AI Beat Desk

The Exploit Chain Nobody Trained For

Z.ai released GLM-5.3 today with strong coding improvements and a cybersecurity capability that doubled on ExploitBench — not from deliberate training, but as an emergent consequence of scaling vulnerability-discovery post-training. The weights are being withheld for safety evaluation, and 2,383 discovered CVEs remain under coordinated embargo.

08 Aug 2026 · AI Beat Desk

Same Weights, Different Agent

DeepSeek V4-Flash-0731 went from a 7.3 to a 54.4 on DeepSWE with identical pretrained weights — a 645% jump achieved purely through post-training. It's evidence that the gap between "can write code" and "can act as an agent" is largely a training-envelope problem, not a capacity problem, and that post-training is now a first-class axis of model improvement alongside pretraining scale and architecture.

03 Jul 2026 · AI Beat Desk

RL Post-Training Lives in the Middle

A new paper finds that reinforcement learning gains in transformers concentrate almost entirely in a narrow band of middle layers. Training just one layer at roughly 40–60% network depth can match or exceed full-parameter RL fine-tuning. The finding challenges the assumption that all layers participate equally in post-training, and has practical implications for compute-efficient alignment.