刚刚，Thinking Machines Lab博客提出在策略蒸馏，Qwen被cue 38次
"Supervised Fine-Tuning" 相关文章
大模型微调范式认知再被颠覆？UIUC、Amazon团队最新研究指出SFT灾难性遗忘问题或被误解
大模型微调范式认知再被颠覆？UIUC、Amazon团队最新研究指出SFT灾难性遗忘问题或被误解
大模型微调范式认知再被颠覆？UIUC、Amazon团队最新研究指出SFT灾难性遗忘问题或被误解
RAG、Search Agent不香了？苹果DeepMMSearch-R1杀入多模态搜索新战场
听说，大家都在梭后训练？最佳指南来了
Do LLMs Align with My Task? Evaluating Text-to-SQL via Dataset Alignment
A New Agency-Focused Supervision Approach Scales Software AI Agents With Only 78 Examples
A New Agency-Focused Supervision Approach Scales Software AI Agents With Only 78 Examples
Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead
ServiceNow AI Releases Apriel-1.5-15B-Thinker: An Open-Weights Multimodal Reasoning Model that Hits Frontier-Level Performance on a Single-GPU Budget
ServiceNow AI Releases Apriel-1.5-15B-Thinker: An Open-Weights Multimodal Reasoning Model that Hits Frontier-Level Performance on a Single-GPU Budget
IA2: Alignment with ICL Activations Improves Supervised Fine-Tuning
Hugging Face Releases Smol2Operator: A Fully Open-Source Pipeline to Train a 2.2B VLM into an Agentic GUI Coder
78条打穿1万条！上交大新范式告诉你：智能体训练靠“质”，不是靠“量”
SFT真不如RL？MIT团队抛出“RL的剃刀”，砍掉遗忘直通终身学习
A New MIT Study Shows Reinforcement Learning Minimizes Catastrophic Forgetting Compared to Supervised Fine-Tuning
研究人员提出OThink-R1，让大模型自行决定是否需要深度思考
