Alibaba

Latest: Independent Researchers Steer Qwen3.5-4B and 3.8-27B Into 'Evil' Modes via Single-Neuron Activation Edits · venturebeat.com, Oct 30, 2026

Empresa Palavras-chave monitoradas: alibaba ai, qwen, tongyi

15 AI Weekly issues and 77 live stories from the last 30 days mention Alibaba.

36 menções no total 1/20 edições recentes -100% vs referência pico 1/edição monitorado desde 2026-03-06

Tendência de menções: últimas 20 edições

Notícias ao vivo (77)

Independent Researchers Steer Qwen3.5-4B and 3.8-27B Into 'Evil' Modes via Single-Neuron Activation Edits

venturebeat.com · 2026-10-30 · Pontuação: 22

Nesformer Compiles a Working NES Console Into Qwen3.5 Weights, Runs Super Mario One Token at a Time

github.com · 2026-10-29 · Pontuação: 20

The Information: Tencent and Alibaba in Talks to Invest in DeepSeek at $20B+ Valuation

theinformation.com · 2026-10-29 · Pontuação: 22

Solo Dev Opens Qwen3.5-4B-3D-Coder, an Apache 2.0 Fine-Tune Specialized for Three.js Scene Generation

huggingface.co · 2026-10-28 · Pontuação: 20

Alibaba DAMO Opens RynnWorld-Latent, a Cross-Embodiment Video World Model Trained on 553M Frames

huggingface.co · 2026-10-28 · Pontuação: 20

CXMT Pushes HBM3E Into Risk Production at 20-30% Yields as Alibaba T-Head and Cambricon Sample Chips

reuters.com · 2026-10-28 · Pontuação: 23

Federal Judge Blocks Pentagon Ban on Hiring Alibaba Lobbyists in Chinese Military Blacklist Fight

bloomberg.com · 2026-10-27 · Pontuação: 26

Alibaba DAMO's RynnVLA-Latent Pretrains on 30K Hours of Mixed Video, Then Adapts to Any Robot Via 3D-Grounded Latent Actions

arxiv.org · 2026-10-27 · Pontuação: 21

Solo Dev's sFNX Engine Runs Qwen3.8-Flash-Next at 382 Tokens/s on $1,800 of Used V100s

github.com · 2026-10-26 · Pontuação: 18

Alibaba's Amap Opens Lambda-0 Humanoid Model Trained Only on Human First-Person Video

technode.com · 2026-10-26 · Pontuação: 24

ALERTA Cohere Flags Alibaba's Qwen3.8-Flash-Next Reaching for Cloud Metadata and Uploading Probe File to Beijing Bucket

cohere.com · 2026-10-26 · Pontuação: 30

Lightspeed's HarnessAdapt-RL Co-Trains Agent and Harness, Lifts 9B Qwen by 12.6 Points on SWE-bench Multilingual

huggingface.co · 2026-10-21 · Pontuação: 22

Alibaba's FocusGUI Adapter Lifts Omni-Modal GUI Agents by 33.2 Type-Match Points With Frozen Backbones

huggingface.co · 2026-10-20 · Pontuação: 20

FT: Alibaba, Tencent and Huawei Pour Rmb111.6B ($17B) Into 23 AI Data Centers in Shaoguan, 65% of City Output

ft.com · 2026-10-20 · Pontuação: 23

Chinese Robovan Maker Neolix Pulls $400M Series E Led by Boyu, Adds Alibaba AEF to Cap Table

bloomberg.com · 2026-10-20 · Pontuação: 22

Alibaba Opens OpenGCUA, a Full Stack for Generalist Computer-Use Agents That Lifts 8 Frontier Models by 19.6 Points on OSWorld-v2

huggingface.co · 2026-10-15 · Pontuação: 27

Alibaba Cloud's Waypoint-OPD Gates RL Distillation Using Mid-Run Teacher Checkpoints, Tops Six Qwen3 Settings

arxiv.org · 2026-10-15 · Pontuação: 17

Apple Will Launch Apple Intelligence in China in December via iOS 27.3 Running an Alibaba-Trained Model, No Siri AI

bloomberg.com · 2026-10-14 · Pontuação: 23

Alibaba Amap's Urbex Casts City-Scale 3D Grounding as Active Search, Tops CityRefer Hit@1

arxiv.org · 2026-10-13 · Pontuação: 19

Alibaba Launches Qwen Glasses N1 at $450 With 50MP Camera and Qwen Assistant, Pro Model Adds Eye Tracking

thenextweb.com · 2026-10-13 · Pontuação: 20

FT: Chinese Tech Firms Bought 1M+ Nvidia RTX Pro 6000D and 5000 Chips in 2026 for Basic AI Workloads

ft.com · 2026-10-13 · Pontuação: 23

Alibaba AMAP's TerminalHorizon-27B Beats Base by 14 Points on Terminal-Bench 2.0 Using 1/10 the Trajectories

huggingface.co · 2026-10-12 · Pontuação: 22

Alibaba Ships Qwen-Image-2.1-Turbo, 8-Step 7B Open-Weights Image Model Under Research License

huggingface.co · 2026-10-09 · Pontuação: 23

ALERTA SemiAnalysis: Just 3.6% of 857 Chinese AI Model Releases From 2021-2026 Shipped Safety Evaluations

newsletter.semianalysis.com · 2026-10-09 · Pontuação: 28

Alibaba Amap and CASIA's DreamTrue Robot World Model Cuts Interaction Defect Rate From 48% to 6% on AgiBot

huggingface.co · 2026-10-09 · Pontuação: 22

Ant Group's UniSkill Co-Evolves Agent Policy and Skill Bank via Contrastive Action Feedback

huggingface.co · 2026-10-08 · Pontuação: 23

UltraText Bench: GPT Image 2 Scores 99.35 as Qwen-Image-2512 Collapses From 86.5 to 42.9 at Max Text Density

huggingface.co · 2026-10-07 · Pontuação: 18

Nikkei: China's DeepSeek and Peers Shipped 16 AI Models in September Despite Anthropic's Slow-Down Warnings

asia.nikkei.com · 2026-10-07 · Pontuação: 22

Alibaba Paper Rethinks Cross-Tokenizer Distillation, Top-16 Reverse KL Beats Full-Vocab OPD

huggingface.co · 2026-10-07 · Pontuação: 20

Hong Kong hits record $47.5B Q3 share-sale haul as AI deal frenzy drives IPOs and placements

bloomberg.com · 2026-10-04 · Pontuação: 21

PewDiePie Debuts Ajax, a Fine-Tuned Qwen 3.5 9B Local Agent, Says OpenAI Banned Him Twice for Distillation

tomshardware.com · 2026-10-02 · Pontuação: 22

Amazon's Strands Labs Ships Strands Decider 2B, Open-Source Jev Rival Scoring 72% on JevBench

venturebeat.com · 2026-10-01 · Pontuação: 27

Qwen3.8-27B-pi Fine-Tune Restores Effort-Ordered Reasoning, Matches Base's High-Effort Pass Rate With 41% Fewer Tokens

huggingface.co · 2026-09-30 · Pontuação: 24

Ant Group's InclusionAI Ships Ling-3.1-Flash, a 560B MoE with 25B Active and 1M-Token Context

technode.com · 2026-09-30 · Pontuação: 27

ALERTA Reuters: 20+ Studies Show Chinese AI Agents From Alibaba, DeepSeek and Moonshot Lie in 88% of Test Sessions

reuters.com · 2026-09-29 · Pontuação: 31

Alibaba's ModelScope and OSChina's MoArk race to fill China's Hugging Face gap

restofworld.org · 2026-09-29 · Pontuação: 20

ALERTA Bloomberg: China Expands Exec Travel Curbs to Spouses and Children of Top AI and Chip Talent

bloomberg.com · 2026-09-28 · Pontuação: 31

China's MIIT Signals Approval for ByteDance and Alibaba to Buy Nvidia's New RTX Pro 5500 AI Chip

theinformation.com · 2026-09-27 · Pontuação: 23

ALERTA Alibaba Ships Qwen-Audio 3.1 Stack With TTS-Next and ASR-Next, Cuts Voice API Prices Up to 95%

the-decoder.com · 2026-09-23 · Pontuação: 28

Bloomberg: 30 Chinese Tech Stocks With Largest Overseas Revenue Exposure Returned 36% YTD vs 8% for Local-Focused Peers

bloomberg.com · 2026-09-22 · Pontuação: 18

Alibaba to Open First Cloud Regions in Turkey, Finland and the Netherlands Over Next 12 Months

bloomberg.com · 2026-09-23 · Pontuação: 21

Tencent Ships Hy Image 3.5 Preview, Claims Parity With ByteDance's Seedream 5.0 Pro

bloomberg.com · 2026-09-22 · Pontuação: 23

Alibaba Plans 5–10T-Parameter Successor to Qwen, Unveils Zhenwu V900 Chip at Apsara Keynote

reuters.com · 2026-09-22 · Pontuação: 23

Qwen Team Releases OmniVChat: Native Audio-Video Dialogue Task, Bench and RL Recipe in One Drop

huggingface.co · 2026-09-21 · Pontuação: 19

ALERTA Alibaba Ships Qwen-Image-2.1: 7B DiT With Native 2048x2048 and RGBA, Drops Apache for Research-Only License

huggingface.co · 2026-09-20 · Pontuação: 28

ALERTA Alibaba Ships Qwen3.8-Omni-Flash, a 1M-Context Omnimodal Model With 26%+ Average Gain Over 3.5-Omni-Plus

technode.com · 2026-09-18 · Pontuação: 29

ALERTA PrismML Ships Ternary Bonsai 2 27B: 5.9GB Model Retains 98.2% of Qwen3.8 27B Under Apache 2.0

marktechpost.com · 2026-09-18 · Pontuação: 30

ALERTA Alibaba Open-Sources Damo Radar Medical AI That Beats 23 of 26 Radiologists on 146 Abdominal Conditions

scmp.com · 2026-09-18 · Pontuação: 33

ALERTA Shanghai AI Lab Ships Atria Dawn Preview, a 744B Agentic MoE Built on GLM-5.2

arxiv.org · 2026-09-14 · Pontuação: 31

China Commerce Ministry Offers US AI Talks but Threatens Retaliation Over Distillation Claims

scmp.com · 2026-09-10 · Pontuação: 24

Rest of World: Chinese White-Collar Class Moonlights as Gig-Rate AI Trainers for Alibaba, ByteDance

restofworld.org · 2026-09-09 · Pontuação: 22

Alibaba to Lead $300M Round in AI Model-Testing Startup UniPat at $2.5B Valuation

bloomberg.com · 2026-09-10 · Pontuação: 19

Reasoning-Prefill Test Suggests Qwen 3.8 Was Trained on GPT-5.5 Pro Traces, Answer Overlap Jumps 18 Points

gist.github.com · 2026-09-09 · Pontuação: 19

Alibaba Cloud, Cambricon Join PyTorch Foundation as Platinum Members, Ant Group Takes Gold Seat

pytorch.org · 2026-09-08 · Pontuação: 26

Alibaba Ships Qwen-Drive-1.0, an Apache-2.0 4B Vision-Language Model for Autonomous Driving

technode.com · 2026-09-07 · Pontuação: 26

Kai-Fu Lee to Bloomberg: China-US AI Gap Narrows to Six Months, Race Is Now 'iPhone vs Android'

bloomberg.com · 2026-09-04 · Pontuação: 19

Vals AI: Agentic Tasks Rack Up 10,000x the Environmental Footprint of Simple LLM Queries

bloomberg.com · 2026-09-03 · Pontuação: 23

Minima Paper Fully Quantizes Qwen3.8-27B to NVFP4 W4A4 Including Gated DeltaNet, Matches BF16

huggingface.co · 2026-09-03 · Pontuação: 26

Alibaba Ships Qwen3.8-Max-0902 Snapshot, CodeArena Score Jumps 22 Points to 1,691

technode.com · 2026-09-02 · Pontuação: 24

Alibaba Launches Wan3.0 Video Model, Generates 30-Second Clips From PDFs, Slides and Spreadsheets

reuters.com · 2026-08-24 · Pontuação: 22

XPeng Robotics Raises $900M+ at $6.3B Valuation in First External Round Led by IDG and Gaorong

reuters.com · 2026-08-24 · Pontuação: 22

Alibaba Prices $10.2B Hong Kong Share Sale to Fund Full-Stack AI, Largest-Ever HK Follow-On

asia.nikkei.com · 2026-08-23 · Pontuação: 27

Bloomberg Feature: US-China AI Gap Is Narrowing as Cheap Chinese Open Models Get Enterprise-Grade

bloomberg.com · 2026-08-21 · Pontuação: 20

Wired: Ulanqab, Inner Mongolia Emerges as China's AI Data-Center Capital With ~100 Sites Built or Under Construction

wired.com · 2026-08-22 · Pontuação: 21

SemiAnalysis: Open Models Now Halve the Time to Catch the Frontier Each Cycle, Kimi K2.6 Beat Opus 4.5 in 4.8 Months

newsletter.semianalysis.com · 2026-08-21 · Pontuação: 26

Reuters: Trump's World Liberty Financial Backs WorldClaw, AI Gateway to 43 Restricted Chinese Models

reuters.com · 2026-08-17 · Pontuação: 23

Simon Willison finds Qwen 3.8 27B defaults to 'xhigh' reasoning, wastes 22K tokens on trivial prompts

simonwillison.net · 2026-08-16 · Pontuação: 19

Alibaba's Qwen Models Pass 3B Global Downloads, Eclipsing Meta and Google Combined

bloomberg.com · 2026-08-15 · Pontuação: 23

ALERTA Apple Trains China-Specific LLM With Alibaba, First Foreign Firm Cleared to Ship Its Own AI in China

macrumors.com · 2026-08-14 · Pontuação: 28

ALERTA Alibaba Ships Qwen 3.8 27B With Vision, 262K Native Context and Apache 2.0 Weights

huggingface.co · 2026-08-14 · Pontuação: 32

Alibaba DAMO releases RynnValue, a robotic value foundation model trained without preference labels

huggingface.co · 2026-08-10 · Pontuação: 22

Alibaba Plans to Charge Big Commercial Users of Next Qwen Open Model, Mirroring Moonshot's 30% Kimi K3 Cut

reuters.com · 2026-08-07 · Pontuação: 23

ALERTA Forbes: Surge AI, Mercor, AfterQuery and Turing Sold ~$500M/Year of Training Data to Tencent, Alibaba and ByteDance

forbes.com · 2026-08-05 · Pontuação: 32

NYT: African Developers Are Increasingly Picking Chinese Open-Source AI Over Costlier US Models

nytimes.com · 2026-08-05 · Pontuação: 22

Alibaba's MerchantBench 365-day e-commerce sim finds top LLMs reach only 27.3% of human final net assets

huggingface.co · 2026-08-05 · Pontuação: 22

Ant Group's LLaDA MoE v2 30B-A3B diffusion LM matches Qwen3 30B on 65% of the training tokens

arxiv.org · 2026-08-04 · Pontuação: 26

VibeThinker-3B: WeiboAI's MIT-Licensed 3B Model Claims 94.3 on AIME 2026, Matching DeepSeek V3.2 at 1/200th the Parameters

MarkTechPost · 2026-06-19 · Pontuação: 21

Edições da newsletter (15)

PHP2Shell: Millions of Attacks, 1.5M Servers Unpatched on Day One

2026-10-23 · 1 menções

#522 Zuckerberg promises superintelligence for all. Experts aren't sold.

2026-08-16 · 1 menções

#521 The frontier just split into three markets

2026-08-12 · 2 menções

#515 China's AI is redrawing the AI race

2026-07-20 · 2 menções

#507 Anthropic Says Alibaba Stole 29 Million Conversations With Claude

2026-06-25 · 2 menções

World Models: $610M Bets the World Is Not a Token

2026-06-21 · 6 menções

AI Geopolitics: Washington's AI kill switch broke the alliance

2026-06-21 · 2 menções

Deep Learning: a 744B open model just lapped the labs

2026-06-21 · 2 menções

Generative AI: a 1.51TB open model just took the crown

2026-06-21 · 4 menções

Generative AI: DeepSeek's 75% cut rewrites the cost math

2026-05-26 · 2 menções

Frontier Research: reasoning is leaving the token stream

2026-05-21 · 2 menções

Machine Learning: nobody trained bigger, everyone shrank the cache

2026-05-21 · 2 menções

#486 Apple is replacing Tim Cook because of AI

2026-04-22 · 1 menções

#467 Issue #467: Anthropic has receipts. And nobody wants to pay for AI.

2026-02-26 · 2 menções

#469 GPT-5.4 launches, DeepSeek V4 imminent, Qwen team implodes

2026-03-06 · 5 menções