Unverified50% confidenceFactExact time
LLaDA-8B adopts large-scale masked diffusion pretraining and approaches autoregressive models of the same scale on instruction-following tasks; Dream-v0-Instruct-7B explores the instruction fine-tuning path of diffusion models in dialogue scenarios.
1
Sources
50%
Confidence
Long-term
Relevance
11/6/2026
First Seen
Sources
Related Entities
Related Claims
Unverified提示词与模型版本存在深度绑定关系，不同版本模型在RLHF训练数据、指令微调策略上的差异会导致相同提示词产生不同输出，这种现象被称为「提示词漂移」69% similarUnverifiedThe autoregressive decoding nature of LLMs inherently favors high-probability sequences, and alignment training such as RLHF further reinforces dominant patterns of human preference, causing generated dialogue strategies to be highly concentrated.69% similarUnverifiedRLHF的人类介入发生在训练阶段并固化为模型权重，属于隐式编码，而表征工程干预在推理时操控内部表征，具有即时性和可逆性68% similarUnverifiedMainstream code large models all adopt a two-stage strategy of "pretraining + instruction fine-tuning," with the second stage using alignment techniques such as RLHF or DPO.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1142352API
curl https://kongchang.com/api/v1/knowledge/claims/1142352MCP
get_claim(id=1142352)