AI
AI
资讯
alishangtian.com
首页
大模型
智能体
开源项目
研究前沿
行业动态
知识库
主题色 · THEME
靛蓝（默认）
极光
落日
薰衣草
海洋
森林
暮橙
石墨
自定义
恢复默认
提交线索
agent
anthropic
openai
huggingface daily papers
ai安全
gpu
jake wharton
it之家
solidot
港股
搜索：
Text-to-SQL
共命中 50 条（服务端检索）
一篇读懂结构化数据 RAG：表格、数据库与
Text
-
to
-
SQL
企业最有价值的知识往往不在文档里，在数据库里。
Text
-
to
-
SQL
让 LLM 把自然语言变成查询语句，但这条路远比「生成一段
SQL
」复杂。本篇拆解结构化数据 RAG 的完整管线：Schema 理解、
SQL
生成、执行验证与错误恢复，以及它与向量 RAG 的本质差异。
一叶一世界
精选
· 原创 · 昨天
阅读 1
·
访客 1
一篇读懂结构化数据 RAG：表格、数据库与
Text
-
to
-
SQL
企业最有价值的知识往往不在文档里，在数据库里。
Text
-
to
-
SQL
让 LLM 把自然语言变成查询语句，但这条路远比「生成一段
SQL
」复杂。本篇拆解结构化数据 RAG 的完整管线：Schema 理解、
SQL
生成、执行验证与错误恢复，以及它与向量 RAG 的本质差异。
一叶一世界
精选
· 原创 · 昨天
阅读 0
·
访客 0
MIRA: A Musical Intent Refinement Agent for Aligning
Text
-
to
-Music Generation with User Intent
Text
-
to
-music systems produce increasingly convincing audio, yet evaluation reveals little about whether the result matc…
智能体
HuggingFace Daily Papers · 10-7
阅读 49
·
访客 47
Post-Training Frontier
Text
-
to
-Image Models by Composing Preference and Rubric Rewards
Recent
text
-
to
-image generation models have achieved remarkable visual quality, but improving them through post-training…
行业动态
HuggingFace Daily Papers · 10-2
阅读 22
·
访客 22
TerraVis:
To
wards Evaluation of World-Grounded Visual Consistency in
Text
-
to
-Image Generation via MLLM Workflows
Recent
text
-
to
-image models have made substantial progress in pho
to
realism, aesthetics, and
text
-image alignment. Yet vi…
大模型
HuggingFace Daily Papers · 10-2
阅读 12
·
访客 12
RenderRank: Learning
to
Rerank
Text
with Compressed Visual
To
kens
Rendering document
text
as images allows vision-language models
to
encode documents as visual
to
kens, which can reduce i…
行业动态
HuggingFace Daily Papers · 9-28
阅读 23
·
访客 23
SpaceXAI Releases Grok Voice Transcribe 2.0: A Speech-
to
-
Text
API Claiming 2x Accuracy Over 1.0 at $0.10 per Hour
SpaceXAI has released Grok Voice Transcribe 2.0, its newest speech-
to
-
text
model for batch and streaming audio. The comp…
行业动态
MarkTechPost · 9-19
阅读 51
·
访客 51
WanPE:
To
wards Cinematic Prompt Enhancement for Modern
Text
-
to
-Video Generation
Video generation begins in
text
space by authoring a cinematic screenplay, then materializes in
to
pixels. As contemporar…
行业动态
HuggingFace Daily Papers · 9-24
阅读 36
·
访客 36
OpenAI launches Decisions API that reduces complex evaluations
to
yes, no, or pick one
Oct 7, 2026 OpenAI announced two updates
to
its API platform.** The new Decisions API evaluates
text
, images, or both, r…
智能体
The Decoder · 10-7
阅读 58
·
访客 57
Google's new Flash TTS models let you design AI voices from scratch using
text
descriptions
Sep 23, 2026 Nano Banana Pro prompted by THE DECODER Key Points Google has released two new
text
-
to
-speech models, Gemin…
大模型
The Decoder · 9-24
阅读 56
·
访客 55
FLEET: From Logits Entropy
to
Enhanced Trajec
to
ries in
Text
Generation
Solutions based on large language models (LLMs) often rely on temperature sampling
to
improve accuracy and stability by …
大模型
HuggingFace Daily Papers · 9-23
阅读 30
·
访客 30
All-in-One Multilingual Scene
Text
Recognition with Script-aware Mixture-of-Experts
Multilingual scene
text
recognition (STR) remains challenging due
to
the scarcity of training data for most languages an…
行业动态
HuggingFace Daily Papers · 9-21
阅读 31
·
访客 31
原生多模态与 Any-
to
-Any：当理解与生成合流进一个模型
GPT-4o 之前的多模态是「拼装」：LLM 外挂视觉编码器、外挂图像生成器。原生多模态把理解与生成合进一个模型——语音进出、图像进出都在同一套权重里。本篇拆解三代多模态架构（外挂拼接、原生统一、any-
to
-any）、早融合与晚融合的技术分野、统一模型的权衡，以及「模态是什么」这个被重新打开的问题。
研究前沿
精选
· 原创 · 2天前
阅读 13
·
访客 13
Block's new AI model Dessa was trained entirely on human speech
to
catch emotions
text
-based models miss
Oct 24, 2026 Block, the company behind Square and Cash App, has unveiled an AI speech model called Dessa.** Block says D…
行业动态
The Decoder · 10-24
阅读 27
·
访客 27
Fish Audio Releases Drama 3, a 35B Open-Weight MoE
Text
-
to
-Speech Model
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态
MarkTechPost · 10-15
阅读 50
·
访客 50
Microsoft AI Releases MAI-Transcribe-2-Streaming: #1 Real-Time Speech-
to
-
Text
Model on Artificial Analysis
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态
MarkTechPost · 10-3
阅读 30
·
访客 30
Microsoft AI releases new transcription and
text
-
to
-speech models for voice agents
Oct 2, 2026 Microsoft AI has released MAI-Transcribe-2-Streaming, a new model for real-time transcription.** Microsoft s…
智能体
The Decoder · 10-2
阅读 62
·
访客 62
Sarvam AI Releases Saaras V4: A Speech-
to
-
Text
Model for All 22 Indian Languages and Global English
Sarvam AI has released Saaras V4, the newest generation of its speech recognition model. It covers all 22 scheduled Indi…
行业动态
MarkTechPost · 9-27
阅读 55
·
访客 55
End-
to
-End Multimodal Data Augmentation and Adversarial Robustness Benchmark with AugLy for Images,
Text
, Audio, and Py
To
rch
In this tu
to
rial, we build a comprehensive multimodal augmentation and robustness workflow with **AugLy** for images, te…
研究前沿
MarkTechPost · 9-26
阅读 41
·
访客 41
Muse Spark Open is Meta's answer
to
Qwen, GLM, and Kimi in the open-weight race
Oct 15, 2026 After its Llama series, Meta is returning
to
open-weight terri
to
ry with Muse Spark Open.** Meta calls it th…
智能体
The Decoder · 10-16
阅读 73
·
访客 73
Figma announces new features
to
define how agents work on its platform
Design platform Figma announced a new Markdown-based edi
to
r
to
help define a company’s design, bring in design and code …
智能体
TechCrunch · 10-15
阅读 76
·
访客 75
An Anthropic AI model sent a false homicide tip
to
Philadelphia police
An Anthropic AI model submitted a false tip about an unsolved murder
to
the Philadelphia police. The AI reportedly submi…
行业动态
TechCrunch · 10-10
阅读 33
·
访客 33
Here are the
to
p AI agents that can live in your
text
messages
Rather than downloading another app, a growing number of agents can simply be
text
ed like an ordinary person. You
text
i…
智能体
TechCrunch · 10-10
阅读 202
·
访客 201
AI access makes people almost entirely unwilling
to
say "I don't know," study finds
Sep 26, 2026 Nano Banana Pro prompted by THE DECODER Researchers ran five experiments with 3,132 participants
to
test wh…
行业动态
The Decoder · 9-27
阅读 49
·
访客 49
Some Supabase cus
to
mers are publicly exposing reams of people’s data
to
the web
Thousands of databases hosted by development platform Supabase are exposing people’s sensitive information
to
the public…
行业动态
TechCrunch · 9-26
阅读 37
·
访客 37
I created an interactive digital avatar of myself — and you can talk
to
it
When Alexandru Voica, head of corporate affairs at the video-generation startup Synthesia, sent me a link this summer
to
…
行业动态
TechCrunch · 9-26
阅读 35
·
访客 35
ElevenLabs’ CEO on margins, IPO timing, and telling cus
to
mers they’re talking
to
a bot
ElevenLabs builds the voice layer of AI, the models that turn
text
in
to
speech that sounds human. Most people encounter …
智能体
TechCrunch · 9-25
阅读 75
·
访客 74
A Coding Guide
to
TypeSafe AI Jev: Typed Decisions, Calibrated Confidence, and Speculative Fan-Out with a System One Model
In this **tu
to
rial**, we work with **Jev**, TypeSafe AI’s first System One model, which does not generate
text
at all: w…
行业动态
MarkTechPost · 9-24
阅读 52
·
访客 51
OpenAI hires Patreon co-founder Sam Yam
to
lead a new Crea
to
r Product division
Sep 23, 2026 OpenAI is building new
to
ols for crea
to
rs as its AI models get better at generating
text
, images, audio, an…
行业动态
The Decoder · 9-23
阅读 23
·
访客 23
CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized
Text
Generation
Adapting large language models
to
individual users remains challenging due
to
the tension between fine-grained personali…
行业动态
HuggingFace Daily Papers · 9-20
阅读 28
·
访客 28
Runway wants
to
turn AI video generation in
to
a live stream you control in real time
Runway wants
to
stream AI video as users prompt it, rather than make them wait for finished clips. The approach builds o…
行业动态
The Decoder · 9-20
阅读 47
·
访客 46
TypeSafe AI Releases Jev: A System One Model That Returns Typed, Calibrated Decisions Instead of
Text
TypeSafe AI released Jev, a System One model that answers typed questions with probabilities instead of generating
text
.…
行业动态
MarkTechPost · 9-20
阅读 54
·
访客 52
Tencent's Gander aims
to
keep talking while it works in the background
Tencent's Gander processes speech, images, and
text
while handling tasks in the background. A "cerebellum" keeps the con…
行业动态
The Decoder · 9-20
阅读 41
·
访客 40
OmniVBench: A Benchmark and Large-Scale Dataset for Omni Reference-
to
-Video Generation
Reference-
to
-video (R2V) generation is evolving
to
ward increasingly general and versatile reference control, giving rise…
研究前沿
HuggingFace Daily Papers · 9-18
阅读 47
·
访客 47
Refinement Is Inherently Editable: Training-Free Prompt-
to
-Prompt Image Editing with Generative Refinement Network
Text
-guided image editing must introduce the requested changes while preserving unrelated source content. Diffusion-base…
行业动态
HuggingFace Daily Papers · 9-17
阅读 41
·
访客 40
Former OpenAI researcher builds an AI model that judges options instead of writing
text
TypeSafe AI, founded by former OpenAI researcher Diogo Almeida, is releasing a model that deliberately generates no
text
…
行业动态
The Decoder · 9-16
阅读 40
·
访客 40
Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning
to
Agents
Reliable confidence estimation is increasingly central
to
the trustworthy deployment of language models: a calibrated es…
智能体
HuggingFace Daily Papers · 9-15
阅读 63
·
访客 61
HypoEvolve: Genetic Algorithms Enable Multi-Agent LLMs
to
Discover Scientific Hypotheses
Scientific agents contribute
to
hypothesis discovery by synthesizing evidence, assessing proposals, and developing new e…
智能体
HuggingFace Daily Papers · 9-14
阅读 70
·
访客 69
Pick Your Poison: Learning
to
Select Poison Sets for Stronger LLM Backdoor Attacks
Backdoor poisoning attacks add poisoned examples
to
otherwise-clean finetuning data, pairing a trigger with a target beh…
大模型
HuggingFace Daily Papers · 9-14
阅读 48
·
访客 48
NVIDIA Brings Real-Time AI
to
Broadcast, Sports and Global Streaming at IBC
At the IBC conference, running Sept. 11-14 in Amsterdam, the creative, technology and business communities are coming
to
…
行业动态
NVIDIA Blog · 9-10
阅读 34
·
访客 34
UniMate: One Unified Model
to
Animate Diverse Skele
to
ns
Recent advances in au
to
matic rigging now deliver animation-ready 3D assets at scale, yet generating the motion
to
drive …
智能体
HuggingFace Daily Papers · 9-4
阅读 50
·
访客 48
To
See a World in a Living Con
text
: Unified Indoor-Outdoor Urban World Generation
Text
-driven 3D generation has advanced rapidly in creating large-scale outdoor environments and detailed indoor scenes, …
行业动态
HuggingFace Daily Papers · 8-29
阅读 30
·
访客 29
Nvidia pushes cloud cus
to
mers
to
cut back on China business
Nov 3, 2026 Nvidia has
to
ld some cloud cus
to
mers
to
take on less business from Chinese companies.** The guidance is mean…
行业动态
The Decoder · 2天前
阅读 1
·
访客 1
NVIDIA Releases Au
to
On
to
logy: An Open Semantic Layer That Grounds AI Agent
SQL
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体
MarkTechPost · 3天前
阅读 0
·
访客 0
父子检索：小块负责找得准，大块负责答得全
向量检索的粒度是个两难：块太小匹配得准但上下文残缺，块太大上下文完整但语义被稀释。父子检索（small-
to
-big）用「小块检索、大块生成」把两个诉求拆开——索引里存的是子块，命中后回父块。本文讲清它的原理、三种常见形态、最小实现与工程取舍。
大模型
精选
· 原创 · 3天前
阅读 14
·
访客 14
From Fragmentation
to
Fusion: A Survey and Roadmap for Omnimodal Embeddings
Embedding models underpin virtually every modern AI system, from search engines
to
recommendation systems. Despite share…
行业动态
HuggingFace Daily Papers · 6天前
阅读 0
·
访客 0
AI delivers ‘zero financial payoff’
to
most companies, Apollo economist says
AI hasn’t benefited the bot
to
m line of any big corporations outside the tech industry, according
to
a note recently publ…
智能体
TechCrunch · 10-26
阅读 60
·
访客 59
Forced
to
pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.
Most AI chatbots won’t outright tell you who
to
vote for, but some of the models behind them will. A new study found tha…
智能体
TechCrunch · 10-24
阅读 60
·
访客 59
Designing an AI serving harness built
to
last
Sponsored Provided byEXL The next competitive advantage in enterprise AI has less
to
do with which model you choose than…
行业动态
MIT Technology Review · 10-23
阅读 18
·
访客 18
NVIDIA Unveils Minerva
to
Bring End-
to
-End Observability for AI Fac
to
ries
Launched
to
day, NVIDIA Minerva is an AIOps platform built
to
help teams find problems earlier and keep AI fac
to
ries prod…
行业动态
NVIDIA Blog · 10-22
阅读 21
·
访客 21