0:000:00
<p>本期的 7 篇论文如下：</p><p>[00:21] 🧠 DeepEyesV2: Toward Agentic Multimodal Model（DeepEyesV2：迈向智能体多模态模型）</p><p>[01:13] 🧭 Visual Spatial Tuning（视觉空间调优）</p><p>[01:54] 🦹 Too Good to be Bad: On the Failure of LLMs to Role-Play Villains（过于完美以致无法邪恶：大语言模型反派角色扮演的失败）</p><p>[02:27] 🧠 Towards Mitigating Hallucinations in Large Vision-Language Models by Refining Textual Embeddings（通过精炼文本嵌入减轻大型视觉-语言模型中的幻觉）</p><p>[03:13] 🪡 Jailbreaking in the Haystack（干草堆中的越狱攻击）</p><p>[03:48] 🎯 CritiCal: Can Critique Help LLM Uncertainty or Confidence Calibration?（CritiCal：语言批判能否校准大模型置信度？）</p><p>[04:23] 🏃 Dense Motion Captioning（密集动作字幕生成）</p><figure><img src="https://image.xyzcdn.net/FsaIHFHhLBTWLGJuTeDK_zAOXkyw.webp"/></figure><p>【关注我们】</p><p>您还可以在以下平台找到我们，获得播客内容以外更多信息</p><p>小红书: AI速递</p>