Unverified50% confidenceSolutionExact time
Common directions for improving pseudo-label quality include confidence filtering, introducing language model rescoring (LM rescoring), and ensembling multiple teacher models.
1
Sources
50%
Confidence
Long-term
Relevance
9/24/2026
First Seen
Sources
Related Entities
Related Claims
Unverified非自回归机器翻译通过知识蒸馏解耦策略使用自回归教师模型生成的伪标签训练，将BLEU分数提升幅度超过10个点，同时推理保持并行解码75% similarUnverified经过指令微调和RLHF训练的模型在遵循格式规范方面表现更好，有时反而会生成格式更完美的虚假引用72% similarUnverified受限解码在保证格式正确性的同时，可能改变模型实际采样分布，导致语义质量的次优选择66% similarUnverifiedThe paper "Diffusion-LM Improves Controllable Text Generation" transfers diffusion models to discrete text for controllable generation and is widely recognized as the seminal work in the direction of diffusion language models.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/948101API
curl https://kongchang.com/api/v1/knowledge/claims/948101MCP
get_claim(id=948101)