无需训练的LLM推理新方法：深度熵引导采样，提升推理效率与质量。
arXiv:2607.09693v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become the dominant paradigm for improving the reasoning capabilitie…
无需额外训练的SSD记忆增强方案，让LLM轻松扩展长上下文记忆。
arXiv:2607.07388v1 Announce Type: cross Abstract: Large Language Models (LLMs) store factual knowledge and domain-specific patterns implicitly in dens…
无需额外训练，利用kNN算法分析LLM内部激活值，实现高效且可动态配置的安全护栏，为AI防护提供新思路。
arXiv:2607.02072v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in domains requiring guardrails to detect uns…
无需训练就能用不确定性指导复杂视觉任务，MLLMs新方法提升多模态模型可靠性。
arXiv:2510.00705v3 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) often struggle with fine-grained perception, such as iden…
块扩散语言模型新突破：无需训练的自推测解码，兼顾质量与速度，少步数下比标准方法更稳健。
arXiv:2603.25702v2 Announce Type: replace Abstract: Block-diffusion language models offer a promising path toward faster-than-autoregressive generatio…
无需训练的大模型抑郁症诊断方法，证据引导的多因素推理大幅提升可靠性
arXiv:2606.10796v1 Announce Type: cross Abstract: Automatic Depression Detection (ADD) from clinical interviews is a pivotal task in computational men…
提出无需训练的关系数据库基础模型，ICML 2026最新研究颠覆传统范式
arXiv:2602.13697v2 Announce Type: replace Abstract: Relational databases (RDBs) contain vast amounts of heterogeneous tabular information that can be …
不做梯度更新也能学习？这篇论文揭示了上下文学习背后的隐式动力学机制
arXiv:2507.16003v4 Announce Type: replace Abstract: One of the most striking features of Large Language Models (LLMs) is their ability to learn in-con…
无需重新训练即可防御多轮越狱攻击，THRD框架精准抓住对话渐进性漏洞，保障LLM安全对话。
arXiv:2606.01738v1 Announce Type: cross Abstract: Multi-turn jailbreak attacks pose a growing threat to LLMs by exploiting conversational dynamics suc…
无需额外训练，用现成大模型就能给数学推理过程打分，性能媲美专用过程奖励模型。
arXiv:2606.01682v1 Announce Type: cross Abstract: Selecting the best response from multiple small-model samples using a stronger scorer is a simple in…
无需训练即可实现零样本/少样本异常检测，智能体模型AnomalyAgent开辟新范式
arXiv:2605.30140v1 Announce Type: new Abstract: Benefiting from generalizability of vision-language models (VLMs) such as CLIP, many zero-/few-shot an…
无需额外训练，仅靠优化提示就能让大模型成为优秀数学导师，LLM教学能力被严重低估。
arXiv:2605.27088v1 Announce Type: cross Abstract: Aligning LLMs for math tutoring typically requires RL-based training with multi-GPU infrastructure. …
多模态大模型无需微调即可作为检索器，FreeRet方法突破传统范式。
arXiv:2509.24621v3 Announce Type: replace Abstract: Multimodal large language models (MLLMs) are emerging as versatile foundations for mixed-modality …
无需训练的病理AI智能体，用惊喜机制引导扫描并共享切片记忆，精准回答全切片图像问题。
arXiv:2605.23559v1 Announce Type: cross Abstract: Whole-slide image visual question answering (WSI-VQA) frames pathology as an extreme-context search …
无需训练即可编排多模态大模型，实现零样本协调多种能力，降低落地门槛
arXiv:2508.10016v4 Announce Type: replace Abstract: Building interactive omni-modal assistants often relies on end-to-end multimodal alignment to fuse…
提出Ghosted Layers，无需训练即可恢复层剪枝后LLM的性能，通过激活对齐解决隐藏状态不匹配问题。
arXiv:2605.15491v1 Announce Type: cross Abstract: Layer pruning removes entire Transformer decoder blocks from large language models, but introduces a…