AllLLMRoboticsStartupsResearchEthicsBusinessHardwareAutonomousMedicineCreative
Topic

#LLM

Curated coverage on «LLM»: releases, research, benchmarks, open-source.

4859 articles-78% vs prev. week296 this week
Subscribe
RELATED:#Business#Hardware#Research#Creative#Startups#Ethics#Autonomous#Robotics#Medicine
Asia's Largest IT Outsourcer Shrugs Off New US Visa Restrictions
LLM·Bloomberg Tech

Asia's Largest IT Outsourcer Shrugs Off New US Visa Restrictions

Oct 9, 2026·2 min
Netflix and Tencent Pay Half of All Sports Rights Fees in Asia, a $6 Billion Market
LLM

Netflix and Tencent Pay Half of All Sports Rights Fees in Asia, a $6 Billion Market

Oct 9, 2026·2 min
SoftBank Seeks Up to $100 Billion from Gulf Investors for AI Bets
LLM

SoftBank Seeks Up to $100 Billion from Gulf Investors for AI Bets

Oct 9, 2026·2 min
Uzum, Uzbekistan's largest private issuer, eyes eurobonds
LLM

Uzum, Uzbekistan's largest private issuer, eyes eurobonds

Oct 9, 2026·2 min
Franklin Templeton CEO Calls AI Investment a Durable Long-Term Opportunity
LLM·Bloomberg Tech

Franklin Templeton CEO Calls AI Investment a Durable Long-Term Opportunity

Franklin Templeton CEO Jenny Johnson told Bloomberg at the Milken Institute Asia Summit 2026 that investing in artificial intelligence remains a durable long-term…

Oct 9, 2026·2 min
Anthropic Banned Being Cruel to Claude, So I Built a Profanity Filter for My Prompts
LLM·Habr AI

Anthropic Banned Being Cruel to Claude, So I Built a Profanity Filter for My Prompts

Anthropic's new rules forbid insulting its models. One developer built Angry Switcher — a Claude Code plugin that intercepts profanity in prompts and cleans them up…

Oct 9, 2026·3 min
OpenAI releases GPT-6.1 Sol Ultrafast on day 4 of its 28-day marathon
LLM·Jiqizhixin (机器之心)

OpenAI releases GPT-6.1 Sol Ultrafast on day 4 of its 28-day marathon

A day after opening GPT-6 to all users, OpenAI shipped a fast GPT-6.1 Sol Ultrafast and updated its model control system. This is day 4 of 28.

Oct 9, 2026·2 min
8B RL-Trained Agent Teaches a Video Model to Look Up Data and Compute Physics
LLM·Jiqizhixin (机器之心)

8B RL-Trained Agent Teaches a Video Model to Look Up Data and Compute Physics

An 8B-parameter agent trained with RL decides on its own when video generation needs search, a character reference, motion simulation, or frame stitching.

Oct 9, 2026·2 min
Top themes
01OpenAI Publishes 722 Math Papers Written by AIOct 9, 2026·2 min02Doubao adds an infinite canvas for creative work and a 2.1 Lite modelOct 9, 2026·2 min03Video Generation With an Expression Gallery Beats Baseline Models on Identity SimilarityOct 9, 2026·2 min04Qwen3-1.7B Fine-Tuned on Raw, Uncaptioned Video Gains 5.1 Points on Image BenchmarksOct 9, 2026·2 min05PCAsplat Cleans Up Gaussian Splat "Floaters" With Local PCA on Three BenchmarksOct 9, 2026·2 min
Researchers Extend DINOv2's CLS Token with Regions of Interest for Image Retrieval
LLM·arXiv cs.CV

Researchers Extend DINOv2's CLS Token with Regions of Interest for Image Retrieval

Oct 9, 2026·2 min
Omni-Diffusion-Distill speeds up multimodal dLLMs by up to 21x: 8 steps instead of 128
LLM·arXiv cs.CV

Omni-Diffusion-Distill speeds up multimodal dLLMs by up to 21x: 8 steps instead of 128

Oct 9, 2026·2 min
Fluid-Gen-Zero hands water physics to a simulator and cuts fluid flow error by 84%
LLM·arXiv cs.CV

Fluid-Gen-Zero hands water physics to a simulator and cuts fluid flow error by 84%

Oct 9, 2026·2 min
LVSPM Beats VGGT at Camera Pose Estimation on 16–256 Frames and Doesn't Break Down on Large Scenes
LLM·arXiv cs.CV

LVSPM Beats VGGT at Camera Pose Estimation on 16–256 Frames and Doesn't Break Down on Large Scenes

Oct 9, 2026·3 min
TopoGPU Computes Persistent Homology 53x Faster on Average
LLM·arXiv cs.CV

TopoGPU Computes Persistent Homology 53x Faster on Average

The authors released TopoGPU, a streaming GPU pipeline for image persistent homology: 53.24x faster than Cubical Ripser on average, and up to 198.01x faster at best.

Oct 9, 2026·2 min
GHARP animates a 3D head in real time: 13x faster on A100
LLM·arXiv cs.CV

GHARP animates a 3D head in real time: 13x faster on A100

GHARP builds an avatar from a few photos and animates it with a lightweight network built for mobile devices: up to 13x faster on A100 and 8x fewer Gaussians.

Oct 9, 2026·2 min
SAVER Teaches Video Models to Reason From a Few Frames, Using 1,250 Examples
LLM·arXiv cs.CV

SAVER Teaches Video Models to Reason From a Few Frames, Using 1,250 Examples

The SAVER framework fine-tunes video LLMs on 1250 examples: dense frames serve as the reference, and the model on sparse frames matches or beats the base Qwen3.5.

Oct 9, 2026·2 min
Distilled Diffusion Models Learn to Forget Unwanted Concepts via DPO
LLM·arXiv cs.CV

Distilled Diffusion Models Learn to Forget Unwanted Concepts via DPO

Oct 9, 2026·2 min
SPLIT-RL Teaches VLMs to See and Reason Separately: +1.4–6.1 pp Over GRPO
LLM·arXiv cs.CV

SPLIT-RL Teaches VLMs to See and Reason Separately: +1.4–6.1 pp Over GRPO

Oct 9, 2026·2 min
Flow matching: velocity scaling cuts FID from 28.0 to 12.2 on ImageNet-256
LLM·arXiv cs.CV

Flow matching: velocity scaling cuts FID from 28.0 to 12.2 on ImageNet-256

Oct 9, 2026·2 min
VICO Trains VLMs on Rewritable Images: Up to +5.0% on New Tasks
LLM·arXiv cs.CV

VICO Trains VLMs on Rewritable Images: Up to +5.0% on New Tasks

Oct 9, 2026·2 min
DOGS Beats 9 Baseline Methods at Logo Generation Across Two Renderers
LLM·arXiv cs.CV

DOGS Beats 9 Baseline Methods at Logo Generation Across Two Renderers

Oct 9, 2026·2 min
LinSlot Teaches a Neural Network to Break a Scene Into Objects and Their Attributes Without Labels
LLM

LinSlot Teaches a Neural Network to Break a Scene Into Objects and Their Attributes Without Labels

Oct 9, 2026·2 min
OcuFlow Removes Eyeglass Glare in One Step and Wins 67.32% of Votes in a Blind Test
LLM

OcuFlow Removes Eyeglass Glare in One Step and Wins 67.32% of Votes in a Blind Test

Oct 9, 2026·3 min
Researchers release VR180 dataset shot on Blackmagic cameras: 1211 stereo samples
LLM

Researchers release VR180 dataset shot on Blackmagic cameras: 1211 stereo samples

Oct 9, 2026·2 min
SLVR Adds 14.2 Points on BLINK Relation to Qwen2.5-VL-7B Without Text CoT
LLM·arXiv cs.CV

SLVR Adds 14.2 Points on BLINK Relation to Qwen2.5-VL-7B Without Text CoT

Researchers propose SLVR: latent reasoning with typed stages. On Qwen2.5-VL-7B, the method added +9.4 on MMVP and +14.2 on BLINK Relation without generating…

Oct 9, 2026·2 min
Google Releases a Workplace AI Agent for Enterprises, but Pricing and Integrations Remain Open Questions
LLM·AI Business

Google Releases a Workplace AI Agent for Enterprises, but Pricing and Integrations Remain Open Questions

Google's agent uses company context for office tasks, but pricing and compatibility with third-party systems remain the key questions for businesses.

Oct 9, 2026·2 min
Related topics
Business 1160Research 516Robotics 178Hardware 106Autonomous 33Startups 9Medicine 8Ethics 8