从第一原理出发，揭示LLM可提取记忆研究中两种相反的效度陷阱，平衡过度提取与不可靠证据的争议。
arXiv:2607.12649v1 Announce Type: new Abstract: Recent work on extractable memorization in LLMs suffers from two contrasting validity problems. Some s…
用NLP对比人类与LLM的语义导航能力，揭秘AI认知机制的新视角。
arXiv:2607.12195v1 Announce Type: new Abstract: Semantic memory retrieval can be conceptualized as navigation through conceptual space. We compared se…
大模型内部发现通用倒数机制，助你在多种任务中精准控制输出长度
arXiv:2607.12279v1 Announce Type: cross Abstract: Writing a sentence of exactly twelve words; ending a DNA sequence at the right codon; formatting an …
中国大模型新贵DeepSeek拟融资15亿美元，估值710亿，2027年或提前IPO。
DeepSeek, the Chinese large language model developer, is said to be preparing for a 2027 IPO debut as it also looks to raise around $1.5 billion in ne…
最新研究揭示情绪诱导如何使大语言模型在序列决策中产生行为偏差，视角独特。
arXiv:2607.12631v1 Announce Type: new Abstract: As Large Language Models (LLMs) are increasingly deployed as autonomous agents in high-stakes domains,…
针对长思维链大语言模型的KV缓存量化新突破，渐进混合精度方案，已被ICLR 2026接收，代码已开源。
arXiv:2505.18610v2 Announce Type: replace Abstract: Recently, significant progress has been made in developing reasoning-capable Large Language Models…
通过合成指令数据扩展预训练规模，突破传统监督训练数据瓶颈的新方法
arXiv:2601.22146v2 Announce Type: replace-cross Abstract: Due to limited supervised training data, large language models (LLMs) are typically pre-trai…
探索大语言模型的“自我反思”能力：元认知研究综述，从理论到前沿机遇全梳理。
arXiv:2607.11881v1 Announce Type: cross Abstract: Metacognition is a foundational component of intelligence critical to effective learning, problem so…
单一向量让大模型同时记住新旧答案，攻克时间事实编辑难题
arXiv:2607.11327v1 Announce Type: cross Abstract: Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts e…
新型多模态检索增强框架让大模型精准理解航拍图像细节，解决细粒度视觉推理难题。
arXiv:2604.17889v2 Announce Type: replace Abstract: Despite recent progress in multimodal large language models (MLLMs), reliable visual question answ…
用大语言模型实现过渡感知的多场景游戏世界生成，让玩家在不同场景间无缝导航。
arXiv:2607.11594v1 Announce Type: new Abstract: Multi-scene navigation (clearing an objective in one bounded space and then crossing a portal into the…
GigaChat Audio 是时间感知音频语言模型，精准理解音频内容与时间关系，提升语音识别与生成质量。
arXiv:2607.10387v1 Announce Type: cross Abstract: Temporal grounding in long recordings remains challenging for audio-conditioned LLMs. We present a t…
利用大模型GPT-4.1深度解析近万条客服对话，将满意度分解为多个维度并验证其有效性，揭示AI在客户体验分析中的潜力。
arXiv:2607.11026v1 Announce Type: new Abstract: We used a large language model (GPT-4.1) to annotate the text of about 9,000 support conversations at …
首个公开评估Claude Fable 5在生物医学挑战问题的表现，直面基准饱和与评估方法可靠性问题。
arXiv:2607.10849v1 Announce Type: new Abstract: Frontier language models are increasingly evaluated on biomedical benchmarks, but two problems undermi…
新研究揭示量化大语言模型的可靠性缩放定律，为部署经济高效且可信赖的AI模型提供理论依据。
arXiv:2607.10855v1 Announce Type: new Abstract: Quantization is a powerful strategy to build capable and resource-efficient large language models (LLM…
首个面向能源领域的大模型多模态数据集，助力LLM在电力、能源管理等场景的应用突破。
arXiv:2607.11459v1 Announce Type: cross Abstract: This paper presents the mAIEnergy dataset, an open-access, multimodal corpus developed to support La…
诊断Video-LLM在长视频中跟踪角色的失败，揭示当前最强开源模型未能真正“观看”视频的局限。
arXiv:2607.11078v1 Announce Type: cross Abstract: Can a Video Large Language Model (Video-LLM) follow one person through a long video, keeping track o…
针对大模型CoT推理中“过度思考”的冗余步骤，用最优停止策略动态剪枝，降本增效不丢精度。
arXiv:2607.11089v1 Announce Type: new Abstract: Large Language Models (LLMs) have achieved remarkable success in complex reasoning tasks through Chain…
新方法STEC通过证据压缩提升开放式多跳问答的深度搜索效率与准确性
arXiv:2607.10795v1 Announce Type: new Abstract: In open-domain multi-hop question answering (QA), LLM-based search agents offer a promising approach t…
Laguerre几何为理解大语言模型中的概念表示提供了精确的数学框架，将概念定义为区域而非单点或方向。
arXiv:2607.10578v1 Announce Type: new Abstract: Existing hypotheses represent a concept in an LLM as a single point, a linear direction, or a Gaussian…