AI资讯日报 - 2025/6/30

👨‍🔬 Danush Khanna, Aditya Kumar Guru, Srivarshinee Sridhar, Zidan Ahmed, Rubhav Bahirwani, Meetu Malhotra, Vinija Jain, Aman Chadha, Amitava Das, Kripabandhu Ghosh

HyperCLOVA X THINK Technical Report

学术论文 ArXiv 重要度: 8

介绍HyperCLOVA X THINK，首个专注于推理的大型语言模型，支持韩语和英语。

👨‍🔬 NAVER Cloud HyperCLOVA X Team

Probabilistic Optimality for Inference-time Scaling

学术论文 ArXiv 重要度: 8

提出概率最优性框架，为LLM推理时缩放提供理论指导。

👨‍🔬 Youkang Wang, Jian Wang, Rubing Chen, Xiao-Yong Wei, Qing Li

Concept-Level AI for Telecom: Moving Beyond Large Language Models

学术论文 ArXiv 重要度: 8

探讨大型概念模型在电信领域的应用，超越大型语言模型的限制。

👨‍🔬 Viswanath Kumarskandpriya, Abdulhalim Dandoush, Abbas Bradai, Ali Belgacem

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

学术论文 ArXiv 重要度: 7

引入自动化LLM速度跑基准，评估AI代理在重现研究成果方面的能力。

👨‍🔬 Bingchen Zhao, Despoina Magka, Minqi Jiang, Xian Li, Roberta Raileanu, Tatiana Shavrina, Jean-Christophe Gagnon-Audet, Kelvin Niu, Shagun Sodhani, Michael Shvartsman, Andrei Lupu, Alisia Lupidi, Edan Toledo, Karen Hambardzumyan, Martin Josifoski, Thomas Foster, Lucia Cipolina-Kun, Abhishek Charnalia, Derek Dunfield, Alexander H. Miller, Oisin Mac Aodha, Jakob Foerster, Yoram Bachrach

Multi-View Contrastive Learning for Robust Domain Adaptation in Medical Time Series Analysis

学术论文 ArXiv 重要度: 7

提出多视图对比学习框架，增强医疗时间序列分析的域适应能力。

👨‍🔬 YongKyung Oh, Alex Bui

Can Video Large Multimodal Models Think Like Doubters-or Double-Down: A Study on Defeasible Video Entailment

学术论文 ArXiv 重要度: 7

研究视频大型多模态模型的抽象推理能力，引入可废止视频蕴含任务。

👨‍🔬 Yue Zhang, Jilei Sun, Yunhui Guo, Vibhav Gogate

Sheaf-Based Decentralized Multimodal Learning for Next-Generation Wireless Communication Systems

学术论文 ArXiv 重要度: 7

提出Sheaf-DMFL框架，利用层理论增强多模态设备间的协作。

👨‍🔬 Abdulmomen Ghalkha, Zhuojun Tian, Chaouki Ben Issaid, Mehdi Bennis

Dehazing Light Microscopy Images with Guided Conditional Flow Matching: finding a sweet spot between fidelity and realism

学术论文 ArXiv 重要度: 6

提出HazeMatching方法，平衡显微镜图像去雾的保真度和真实感。

👨‍🔬 Anirban Ray, Ashesh, Florian Jug

Towards Distributed Neural Architectures

学术论文 ArXiv 重要度: 6

探索分布式神经架构，通过模块和路由器实现灵活的计算路径。

👨‍🔬 Aditya Cowsik, Tianyu He, Andrey Gromov

From Ground to Air: Noise Robustness in Vision Transformers and CNNs for Event-Based Vehicle Classification with Potential UAV Applications

学术论文 ArXiv 重要度: 6

评估事件相机在噪声环境下的性能，探索UAV应用潜力。

👨‍🔬 Nouf Almesafri, Hector Figueiredo, Miguel Arana-Catania

🤖 AI资讯日报

📊 今日趋势总结

Ask HN: Is the rate of progress in AI exponential?

Ask HN: What's the pain using current AI algorithms?

50% Cheaper GPUs for cloud-computing / Saving devs 50% compared to AWS

Ask HN: Anyone concerned about NYC Local Law 144?

NLP, AI, ML, bots – a passing trend or much more? What's your take on this?

Common Lisp + Machine Learning Internship at Google (Mountain View, CA)

Ask HN: Dipping my toes with artificial intelligence and what to expect? (CS)

The AI Crackpot Index

Bioinformatician

Ask HN: Thoughts on grad school? (CS PhD)

Show HN: Startup Raising capital through Book Sales

The Next Bill Gates or Albert Einstein in AI “Chris Clark” – Yourobot

onnx/onnx

tensorflow/tfjs

EthicalML/awesome-production-machine-learning

Unity-Technologies/ml-agents

ShusenTang/Dive-into-DL-PyTorch

nndl/nndl.github.io

jantic/DeOldify

e2b-dev/awesome-ai-agents

afshinea/stanford-cs-229-machine-learning

iperov/DeepFaceLab

Zeyi-Lin/HivisionIDPhotos

amark/gun

CLoVE: Personalized Federated Learning through Clustering of Loss Vector Embeddings

QuickSilver -- Speeding up LLM Inference through Dynamic Token Halting, KV Skipping, Contextual Token Fusion, and Adaptive Matryoshka Quantization

HyperCLOVA X THINK Technical Report

Probabilistic Optimality for Inference-time Scaling

Concept-Level AI for Telecom: Moving Beyond Large Language Models

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

Multi-View Contrastive Learning for Robust Domain Adaptation in Medical Time Series Analysis

Can Video Large Multimodal Models Think Like Doubters-or Double-Down: A Study on Defeasible Video Entailment

Sheaf-Based Decentralized Multimodal Learning for Next-Generation Wireless Communication Systems

Dehazing Light Microscopy Images with Guided Conditional Flow Matching: finding a sweet spot between fidelity and realism

Towards Distributed Neural Architectures

From Ground to Air: Noise Robustness in Vision Transformers and CNNs for Event-Based Vehicle Classification with Potential UAV Applications

📅 历史日报目录