Hugging Face Blog发布于 08/21 08:00
Measuring benchmark optimization in speech recognition
（翻译）衡量语音识别中的基准优化
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
arXiv cs.AI发布于 08/26 12:00
Function-Level Execution Feedback for Code Preference Optimization
（翻译）函数级执行反馈用于代码偏好优化
Abstract page for arXiv paper 2608.23632: Function-Level Execution Feedback for Code Preference Optimization
arXiv cs.AI发布于 08/26 12:00
Generating Biomedical Fact-Checking Reports with RL-Enhanced Agentic Search
（翻译）利用强化学习增强的智能体搜索生成生物医学事实核查报告
Abstract page for arXiv paper 2608.23811: Generating Biomedical Fact-Checking Reports with RL-Enhanced Agentic Search
Hugging Face Blog发布于 08/28 08:00
The Open ASR Leaderboard Adds Its First Global South Language
（翻译）Open ASR 排行榜新增首个全球南方语言
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
人人都是产品经理发布于 08/30 10:15
AI产品成功经验分享：AI质检从NLP到大模型
从NLP时代到大模型时代，AI质检的落地成本骤降十倍，准确率却从半年苦练的80%跃升至开箱即达的98%。本文通过两个真实案例，对比不同技术路线下的架构、训练与成本差异，揭示技术革命如何将AI质检从大公司专属的奢侈品，变为中小企业随手可用的生产力工具。
arXiv cs.AI发布于 08/31 12:00
Retrieving Relations, Detecting Fallacies: A RAG Approach to Political Debate Analysis
（翻译）检索关系，检测谬误：一种用于政治辩论分析的RAG方法
Abstract page for arXiv paper 2608.27471: Retrieving Relations, Detecting Fallacies: A RAG Approach to Political Debate Analysis
arXiv cs.AI发布于 08/31 12:00
Thinking Costs Tokens: When More Structure is Worth the Price
（翻译）思考消耗令牌：更多结构何时物有所值
Abstract page for arXiv paper 2608.27506: Thinking Costs Tokens: When More Structure is Worth the Price
arXiv cs.AI发布于 08/31 12:00
Nemotron 3.5 Content Safety Moderator: A Compact Multimodal, Multilingual, and Reasoning Enabled Content Safety Moderator
（翻译）Nemotron 3.5 内容安全审核器：一款紧凑的多模态、多语言且支持推理的内容安全审核器
Abstract page for arXiv paper 2608.27548: Nemotron 3.5 Content Safety Moderator: A Compact Multimodal, Multilingual, and Reasoning Enabled Content Safety Moderator
Hugging Face Blog发布于 09/02 05:39
BenchMIRT: What are LLM benchmarks actually measuring?
（翻译）BenchMIRT：LLM 基准测试究竟在测量什么？
A Blog post by Ai2 on Hugging Face
arXiv cs.AI发布于 09/02 12:00
Incremental Risk Assessment of Progressive Elder Financial Scams via Instruction-Tuned Small Language Models
（翻译）基于指令微调小语言模型的渐进式老年人金融诈骗增量风险评估
Abstract page for arXiv paper 2609.00005: Incremental Risk Assessment of Progressive Elder Financial Scams via Instruction-Tuned Small Language Models
arXiv cs.AI发布于 09/02 12:00
Long-Horizon State Tracking in LLMs: Executing MD5 through a Deep Sequence of Dependent Tool Calls
（翻译）大语言模型中的长程状态追踪：通过深层依赖工具调用序列执行 MD5
Abstract page for arXiv paper 2609.00012: Long-Horizon State Tracking in LLMs: Executing MD5 through a Deep Sequence of Dependent Tool Calls
arXiv cs.AI发布于 09/02 12:00
LLM-Driven Autonomous Vehicles Inherit Human Driver Biases in Pedestrian Yielding: Results and Implications From A New Benchmark
（翻译）LLM驱动的自动驾驶汽车在行人让行中继承人类驾驶员的偏见：来自新基准的结果与启示
Abstract page for arXiv paper 2609.00192: LLM-Driven Autonomous Vehicles Inherit Human Driver Biases in Pedestrian Yielding: Results and Implications From A New Benchmark
InfoQ 中文发布于 09/09 01:06
Cohere 发布 Parse 5，可以从复杂的文档中高效地提取多模态信息
Cohere 正式发布 Parse 5（parse-v5.0），这是一款专有多模态基础模型，专为解决开发者长期面临的难题——从复杂的企业文档中提取结构化数据——而设计。
arXiv cs.AI发布于 09/14 12:00
GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting
（翻译）GLARE：面向社会动态预测的对抗奖励估计生成学习
Abstract page for arXiv paper 2609.12165: GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting
arXiv cs.AI发布于 09/14 12:00
WinSyn: An Automated Pipeline for Realistic Enterprise Question-Answering Evaluation
（翻译）WinSyn：面向真实企业问答评估的自动化流水线
Abstract page for arXiv paper 2609.12171: WinSyn: An Automated Pipeline for Realistic Enterprise Question-Answering Evaluation