用于商业果园早期绿色果实解剖结构分类的轻量级多模态视觉语言框架
该研究提出了一种面向商业果园的轻量级多模态视觉语言框架，用于早期绿色苹果幼果的解剖结构分类。其识别目标包括花萼、幼果主体和果梗等关键结构，这些信息对机器人疏果、作物负载管理及其他精准果园作业十分重要。原文摘要指出，准确识别这些早期果实结构是实现自动化果园管理的基础。所提供原文未包含模型架构、实验数据或性能对比等进一步细节。
独立 AI 资讯库
AI 前沿动态、原文归档与媒体镜像
从自动抓取数据库实时读取，适合快速浏览、检索和后续扩展成完整资讯站。
12005资讯
11873正文归档
75817媒体
融合感知视觉专家与多模态大语言模型，实现可解释的植物病害诊断：从基准图像到真实世界机器人田间验证
研究团队提出混合分层多智能体框架（H$^{2}$MAF），在决策层融合 EfficientNet-B3 和 ConvNeXt-Tiny，并由 Gemma 4 E4B、Qwen3.5 4B 等开源权重多模态大语言模型进行语义仲裁。该系统基于结构化 JSON 证据输出可解释诊断、风险等级、治疗紧急程度和财务风险评估，并在 PlantDoc 及两套康奈尔大学机器人田间数据集上完成测试。在 PlantDoc 上，Gemma 将准确率从 63.9% 提升至 68.5%，在 CNN 预测冲突的 41.7% 样本中提升 7.6 个百分点；在康奈尔田间数据集上的准确率分别达到 99.3% 和 98.9%。研究表明，多模态大语言模型可在视觉模型意见冲突时提供有效仲裁，但风险评估能力高度依赖校准：Gemma 的关键风险误差较低，而 Qwen 存在较明显的风险过度标记问题。
网球多模态伤病风险预测
该研究探索利用多模态数据预测网球运动员的伤病风险。论文指出，机器学习已显著促进运动表现和伤病风险预测，但现有研究大多依赖主观观察与专家评估，因此效果受到限制。对于网球等运动项目，更丰富、客观的多模态数据有望提升伤病风险评估的准确性，并为教练、运动队医和开发者构建预防性分析工具提供参考。
Understanding the Energy Scaling of Large Language Model Inference Across Context Lengths and Attention Architectures
arXiv:2608.25096v1 Announce Type: new Abstract: The growing adoption of large language models (LLMs) has raised increasing concerns about the energy consumption and environmental impact of inference. This paper presents a systematic empirical study of decode-phase energy consumption across represen
MSR-IVA: Masked Structural Residual Independent Vector Analysis for State-Aware Fusion of Structural MRI and Dynamic Functional Network Connectivity
arXiv:2608.24978v1 Announce Type: new Abstract: Multimodal fusion of structural MRI (sMRI) and dynamic functional network connectivity (dFNC) can reveal how brain structure relates to changing functional states. When the same structural latent representation is coupled with multiple states, applyin
AFDBench: A Reasoning-First AI Scientist for NationalWeather Service Forecast Discussions
arXiv:2608.24954v1 Announce Type: new Abstract: Large language models (LLMs) hallucinate numerical values when generating high-stakes meteorological text, posing risks for weather communication. We present AFDBench, an AI meteorologist that generates professional Area Forecast Discussions (AFDs) by
Demystifying Reinforcement Learning Post-Training of Language Models
arXiv:2608.24949v1 Announce Type: new Abstract: Reinforcement learning (RL) post-training has emerged as a powerful framework for enhancing the capabilities of large language models (LLMs), enabling impressive reasoning, math, and coding capabilities. Yet for many researchers and practitioners, the
CAT-GS: Balanced Multimodal Learning via Calibrated Gating and Fusion Surgery
arXiv:2608.24947v1 Announce Type: new Abstract: End-to-end training of multimodal neural networks often exhibits unstable neural dynamics characterized by three coupled failure modes that degrade learning: (i) modality imbalance, where one branch dominates gradient-based optimization; (ii) unstable
MacroAgent: Regularity-Aware Macro Legalization with LLM-Agent-Designed Contour Algorithms
arXiv:2608.24946v1 Announce Type: new Abstract: Macros constitute a large part of the core area in modern very large-scale integration (VLSI) designs. Moreover, macro positions have a significant impact on the final quality of result (QoR), and macro legalization is typically the final step in dete
GreenLeaf Law Embed Tiny: A Compact Embedding Model for Legal Domain Retrieval
arXiv:2608.24936v1 Announce Type: new Abstract: We present GreenLeaf Law Embed Tiny, a 0.6B parameter embedding model for legal domain retrieval. GreenLeaf-Tiny achieves 75.11% on the Massive Legal Embedding Benchmark (MLEB) and 64.38% on MTEB(Law, v1),demonstrating competitive performance among mo
VietAIDetector: An Open-Source Zero-Shot Detector for Vietnamese AI-Generated Text
arXiv:2608.25478v1 Announce Type: new Abstract: In recent years, distinguishing between AI-generated text and human-written text has remained a challenge. In this paper, we introduce VietAIDetector, an open-source tool designed specifically for detecting Vietnamese AI-generated text. It allows user
MathAdv: What Theorem Provers Know, Reason, Formalize, and Generalize
arXiv:2608.25449v1 Announce Type: new Abstract: Formal theorem proving enables machine-verifiable evaluation of mathematical reasoning, yet existing benchmarks often emphasize aggregate proof accuracy, concentrate on a narrow range of mathematics, and provide limited evidence of robustness to equiv
DCGC: Draft-Conditioned Global Correction for Complex Reasoning with Masked Diffusion Models
arXiv:2608.25428v1 Announce Type: new Abstract: Correcting flawed reasoning traces remains a significant challenge for Large Language Models (LLMs), whose autoregressive generation can propagate early mistakes into subsequent reasoning. We introduce DCGC, a Masked Diffusion Model (MDM) framework fo
OmniPhys: A Unified Multimodal Benchmark for Physics Understanding and Generation from Chinese Educational Corpora
arXiv:2608.25398v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have demonstrated strong abilities in solving diverse visual and textual reasoning tasks. However, their development in the physics domain is significantly hindered by the lack of a comprehensive benchmark. To
Adaptive Triggering for Bias Correction in LLM Reasoning
arXiv:2608.25379v1 Announce Type: new Abstract: Chain-of-thought prompting can expose and amplify demographic stereotypes within an LLM's intermediate reasoning and create a failure mode that final-answer debiasing alone cannot address. Mitigating such bias during generation presents a fundamental
Routed Graph Handoff: Adaptive Format Selection for Multi-Agent LLM Delegation
arXiv:2608.25277v1 Announce Type: new Abstract: Multi-agent LLM systems coordinate through natural-language messages that consume 40--60\% of their token budget. Replacing these with structured graphs reduces cost but fails on tasks requiring adaptive reasoning. We propose \textbf{Routed Graph Hand
From Memorization to Absorption: Mixed-Policy RL for Continual Knowledge Injection
arXiv:2608.25243v1 Announce Type: new Abstract: Continual knowledge injection is essential for keeping large language models up-to-date in a fast-evolving world. Existing methods rely on supervised fine-tuning (SFT), which memorizes injected facts in their training format but fails to generalize ac
Belief Cascades Drive Persuasion in LLM Agent Networks
arXiv:2608.25152v1 Announce Type: new Abstract: Multi-agent LLM systems increasingly debate answers, coordinate research, simulate users, and mediate information flows, making agent-to-agent persuasion a basic but undermeasured capability. We introduce a controlled testbed for studying how goal-dir
SelfGraphRAG: Bridging the Supervision Gap in Graph-Based RAG with Synthetic QA Generation
arXiv:2608.25123v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) improves large language models by incorporating external knowledge without retraining, but existing methods often underuse the relational structure encoded in knowledge graphs. Graph-based RAG can capture entity re
Less can be More: Relieving RAG Bottlenecks via Evidence Frontloading and Pressure-Adaptive Budgeting
arXiv:2608.25115v1 Announce Type: new Abstract: Existing methods for improving Retrieval-Augmented Generation (RAG) efficiency mainly optimize downstream LLM generation, such as context compression or serving optimization. However, RAG is an end-to-end system, and its bottleneck can shift between u
MTDiag: A Multi-Turn Diagnostic Dataset Towards Clinically Meaningful LLM Evaluation
arXiv:2608.25085v1 Announce Type: new Abstract: Clinical diagnosis is fundamentally interactive and incremental, yet the dominant paradigm for evaluating Large Language Models (LLMs) in medicine remains static QA benchmarks or template-based dialogues. These benchmarks say little about whether a mo
HealthBench-Psych: A Mental Health Subset of OpenAI's HealthBench
arXiv:2608.25071v1 Announce Type: new Abstract: General-purpose health benchmarks increasingly anchor claims about LLM medical performance, but they are not always resolved by clinical specialty, making domain-specific performance hard to isolate. Mental health is of acute public-health concern as
Padamitra: Grounded Glossary Generation for Classical Sanskrit
arXiv:2608.25038v1 Announce Type: new Abstract: We introduce grounded glossary generation, a structured task requiring models to recover semantically meaningful Sanskrit phrases and produce translation-grounded meanings from a sloka-translation pair, formalizing the traditional patha commentary pra
A Primer on Computational Semantics for Artificial Intelligence Systems
arXiv:2608.25022v1 Announce Type: new Abstract: As people adopt transformer-based language models (e.g., ChatGPT and Gemini) for an increasing number of use-cases, it is important to know how such models learn and represent the meaning of the language, and to be more informed about what language is
The Imperfective Paradox Is Not Necessarily in Large Language Models: A Benchmark Failure Before a Model Failure
arXiv:2608.25005v1 Announce Type: new Abstract: The imperfective paradox provides a useful test of compositional semantic analysis. Recent work constructs an NLI benchmark and reports that models frequently infer completed telic events from progressive descriptions, attributing this behavior to a T
自动化生成连贯的长篇视频
视频扩散模型已能在数秒内生成高保真短片，但要将其扩展为连贯的长篇叙事，仍面临语义漂移、级联故障、特征漂移和内容坍塌等问题。Google 研究团队推出 AI 视频联合导演（AI video co-director）框架，在 Gemini 和 Veo 之上构建多智能体编排层，将长视频生成视为全局优化与世界状态追踪问题。该体系包含 Co-Director、CANVAS、A²RD 和 VQQA 等框架，可自动完成多模型提示、镜头串联和闭环视觉优化。评估显示，该框架提升了多镜头叙事一致性与角色持续性，能够生成分钟级视频，并减少视觉漂移和流水线错误传播。
视觉语言模型看似存在能力上限，实则是读出方式的限制
研究指出，视觉语言模型基准测试中用于呈现答案选项的约定——如字母、颜色名称或像素坐标——并非中性的评测媒介，可能测到的是模型的读出限制而非真实能力边界。在 200 张 COCO 图像上，Qwen3-VL-4B 以英文位置描述作答时的目标定位准确率为 68.5%，而改用相同位置的像素坐标后降至 20.0%，随机水平为 11.1%。不同答案格式还会改变模型间的胜负关系：原本在英文名称条件下打平的两款模型，在两种坐标系统下分别出现 39 分和 54 分、方向相反的差距。研究建议开发者和评测者将答案表示法视为关键实验变量，并检查评分器、固定词汇表与模型输出形式之间是否存在不兼容。
视觉语言模型能分析以人为中心的视频吗？模型能力与人机协作工作流映射
这篇 arXiv 预印本探讨视觉语言模型（VLM）是否能够用于分析以人为中心的视频内容。论文指出，视频记录了人类行为、互动及其所处情境，是理解个体和开展以人为中心研究的重要证据。研究聚焦于梳理模型能力，并探索人类与 AI 协作的视频分析工作流。该议题对需要处理行为观察、访谈录像、现场记录等视频材料的研究者和开发者具有参考意义。
面向成本高效空间转录组学的主动位点选择基准评测
该论文以 arXiv:2609.27208v1 形式发布，研究空间转录组学中的主动位点选择问题。空间转录组学可在组织上下文中测量基因表达，但高密度捕获网格成本较高，并可能重复采样形态相似的区域。论文指出，现有多数主动学习策略主要为类别标签和独立样本场景设计，因此需要评估其在空间转录组学任务中的适用性。该研究有望为开发者和研究人员选择更具成本效率的空间采样策略提供基准参考。
多样性设计：面向基于原型可解释性的架构约束
该论文研究基于原型的神经网络可解释性问题。此类模型通过基于案例的推理提供内在可解释性，但其原型往往会收敛至冗余特征，难以覆盖多样化的语义部件。论文聚焦于通过架构约束改善原型的多样性与可解释性，这对需要可靠、可审查模型决策依据的开发者和行业应用具有意义。