Unverified50% confidenceFactExact time
大模型在约千亿参数规模附近突然展现出思维链推理、多步数学解题等此前小模型不具备的能力
1
Sources
50%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
Getting Started with AI Agents: A Deep Dive into the Perception-Decision-Action Loop
bilibiliAI大模型学习圈7/9/2026
Related Claims
Unverified思维链推理的效果随模型规模呈现涌现性，在较小模型上几乎无效，在足够大的模型上才能显著提升推理准确率82% similarUnverifiedResearchers found that when model parameters exceed a threshold of approximately 10 billion, models exhibit a nonlinear emergent capability leap in tasks such as logical reasoning and multi-step planning.78% similarUnverifiedWhen model scale crosses from billions of parameters to the hundreds-of-billions level, non-linear emergent capability jumps manifest on code tasks.76% similarUnverifiedIn AI decision-making scenarios dominated by large language models, the model's reasoning process is distributed across the activation states of billions of parameters, making it difficult to establish a complete causal chain from input to output.76% similarUnverifiedLarge language models perform poorly on math and logic problems requiring multi-step reasoning, as models tend to 'jump' directly to an answer rather than going through intermediate reasoning steps.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/478526API
curl https://kongchang.com/api/v1/knowledge/claims/478526MCP
get_claim(id=478526)