给LLM判卷装上“安全气囊”：不确定时检索或弃权，还带可证明风险保证。
arXiv:2608.17994v1 Announce Type: new Abstract: Using LLMs as judges has become standard practice for evaluating model outputs at scale. This is parti…
深入剖析免训练低秩压缩中校准与截断误差的传播机制，为LLM高效压缩提供理论支撑。
arXiv:2608.08506v1 Announce Type: new Abstract: Training-free low-rank compression frameworks have been gaining prominence for LLM compression given t…
用结构化记忆给LLM“降噪”，破解过度个性化难题，AI调校新思路。
arXiv:2608.08300v1 Announce Type: new Abstract: Conversational assistants increasingly rely on persistent long-term memory to personalize responses ac…
针对微调后CLIP模型的分布漂移，提出利用图像-文本对齐的漂移感知校准方法，提升跨领域零样本性能。
arXiv:2501.19060v4 Announce Type: replace-cross Abstract: Vision-language models (VLMs), such as CLIP, adapt effectively to downstream tasks through p…
用大语言模型驱动代理agent替代传统校准方法，实现灰盒仿真模型的精准校准，打开科学计算新思路。
arXiv:2607.18308v1 Announce Type: cross Abstract: Calibration of grey-box simulation models is a constrained optimization problem in which model evalu…
研究揭示LLM评估者没有参考答案时评分偏高，需警惕开源模型自评偏差
arXiv:2607.12885v1 Announce Type: new Abstract: LLM judges are increasingly being used to evaluate open-ended model responses, often in no-reference s…
系统解剖LLM中不确定性的来源与量化方法，为可信AI提供理论基础。
arXiv:2603.24967v2 Announce Type: replace Abstract: Understanding why a large language model (LLM) is uncertain about the response is important for th…
大模型在不确定时学会说“不知道”，选择性预测校准方法，提升安全性与可靠性。
arXiv:2607.03528v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as critical decision-making components in hig…
多模态模型如何自知未知？CoMet把不确定性拆解为上下文与多义性，为可靠AI提供新视角。
arXiv:2606.32012v1 Announce Type: new Abstract: Uncertainty estimation has been a long-standing challenge in AI models; it amounts to "knowing what yo…
探针法如何量化LLM的不确定性？这篇研究从信号到迁移层层拆解，为模型校准提供新的因子化视角。
arXiv:2606.27679v1 Announce Type: cross Abstract: Probe-based uncertainty estimation (UE) has emerged as a prominent approach to detect hallucinations…
长文本生成中，模型何时在“胡编”？这项细粒度不确定性量化比较研究，为幻觉检测提供新视角。
arXiv:2602.17431v2 Announce Type: replace-cross Abstract: Uncertainty quantification has emerged as an effective approach to closed-book hallucination…
多模态大模型当评委也有偏见？论文揭示文化模糊下校准失败的关键漏洞。
arXiv:2606.20676v1 Announce Type: cross Abstract: MLLM-as-a-Judge is conventionally validated by agreement with human annotations, but this metric is …
微调大模型检测系统漏洞，看似精准实则可能只是校准错觉，揭示能力边界的关键研究。
arXiv:2606.20502v1 Announce Type: cross Abstract: Whether LLMs scoring well on vulnerability benchmarks genuinely reason about security or merely patt…
思维链推理预算越高，大模型反而越“迷之自信”？新研究揭示校准漂移现象与CABStop缓解方案
arXiv:2606.11211v1 Announce Type: cross Abstract: The ability of large language models (LLMs) to express calibrated uncertainty is important for safe …
新协议ACUTE通过激活操作提升语言模型的校准度、实用性和可信度，为AI安全提供新思路。
arXiv:2606.07822v1 Announce Type: cross Abstract: As language models improve and become increasingly deployed to solve a variety of tasks, trustworthi…
破解大模型“会做却做错”谜题：多项选择题中知识-预测差距的桥接方法，被ICML 2026接收，理论突破与实践价值兼备。
arXiv:2509.23782v4 Announce Type: replace Abstract: While large language models (LLMs) perform strongly on diverse tasks, their trustworthiness is lim…
LLM裁判小组校准：有限预算下低维叠加器vs联合表，谁更优？
arXiv:2606.01034v1 Announce Type: new Abstract: We study when LLM judge panels should be calibrated with low-dimensional stackers versus joint output …
多模态大模型当评委也会“看走眼”？本文提出感知扰动+奖励建模来纠正偏见，实验数据扎实。
arXiv:2606.02578v1 Announce Type: cross Abstract: Recent multimodal large language models have demonstrated strong reasoning ability, yet their reliab…
论文提出LLMs中“虚假不动点”现象，揭示模型在自我反馈循环中错误强化导致稳定失校准，并引入表征压缩机制提供新解释。
arXiv:2510.14925v4 Announce Type: replace Abstract: High-confidence errors in large language models are often treated as fragile failures. We study an…
从密度比视角重新定义后验学习推迟策略，为AI与专家协同提供新理论工具
arXiv:2605.19557v1 Announce Type: cross Abstract: We study post-hoc Learning to Defer (L2D) through the lens of ideal distributions: divergence-regula…