LLM on 时影https://huggingaha.github.io/topics/llm/Recent content in LLM on 时影Hugo -- gohugo.iozh-cnhuggingaha@gmail.com (时影)huggingaha@gmail.com (时影)© 2026 时影Wed, 01 Oct 2025 00:00:00 +0000LoRA：无憾之选https://huggingaha.github.io/posts/lora-without-regret/Wed, 01 Oct 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/lora-without-regret/MBTI-in-Thoughts：基于 MBTI 的 LLM 行为调控与结构化思维框架https://huggingaha.github.io/posts/mbti-in-thoughts/Tue, 30 Sep 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/mbti-in-thoughts/Small Leak Can Sink a Great Ship—Boost RL Training on MoE with 𝑰𝒄𝒆𝑷𝒐𝒑!https://huggingaha.github.io/posts/small-leak-can-sink-a-great-ship/Tue, 23 Sep 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/small-leak-can-sink-a-great-ship/击败 LLM 推理中的非确定性https://huggingaha.github.io/posts/defeating-nondeterminism-in-llm-inference/Sun, 14 Sep 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/defeating-nondeterminism-in-llm-inference/Why Language Models Hallucinate——解构 LLM 幻觉https://huggingaha.github.io/posts/why-language-models-hallucinate/Sun, 07 Sep 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/why-language-models-hallucinate/LLM 幻觉不可避免？https://huggingaha.github.io/posts/llm-hallucinations-taxonomy/Sat, 16 Aug 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/llm-hallucinations-taxonomy/上下文学习中的“位置效应”https://huggingaha.github.io/posts/bicl-prompt-positional-bias/Fri, 01 Aug 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/bicl-prompt-positional-bias/GSPO：组序列策略优化https://huggingaha.github.io/posts/gspo/Mon, 28 Jul 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/gspo/kimi-k2技术报告https://huggingaha.github.io/posts/kimi-k2-technical-report/Tue, 22 Jul 2025 00:00:00 +0000huggingaha@gmail.com (时影)https://huggingaha.github.io/posts/kimi-k2-technical-report/