Unverified50% confidenceFactExact time
One of the fundamental reasons the industry has actively explored alternative paths such as post-training, reinforcement learning, and inference-time scaling in recent years is the diminishing marginal returns of pre-training scaling laws.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/9/2026
First Seen
Valid until: 1/7/2027
Sources
Related Entities
Related Claims
UnverifiedAfter large models entered the post-pretraining era, the capability gains from simply stacking parameters have become increasingly diminishing, and vendors have turned to the post-training phase in search of breakthroughs.76% similarUnverifiedThe optimal guidance weight formula reveals that external guidance should be trusted more in the early stages of training when the model's own variance is large, and that the guidance weight should be gradually reduced as training progresses, providing a theoretical basis for guidance annealing strategies.73% similarUnverifiedAs the scale of pretrained models increases, many previously effective explicit knowledge injection techniques have gradually become ineffective or even produced negative effects.72% similarUnverified不同模型在抗诱导能力上的差异很大程度上源于后训练阶段（post-training）的策略差异71% similarUnverified后训练的成熟度直接决定模型的token效率，成熟的后训练能减少无效推理token的消耗71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/991101API
curl https://kongchang.com/api/v1/knowledge/claims/991101MCP
get_claim(id=991101)