From Pixels to Words Towards Native Vision-Language Primitives at Scale
@_akhaliq
-
Agentic Entropy-Balanced Policy Optimization Framework
By
–
Agentic Entropy-Balanced Policy Optimization
-
Multi-Turn Text-to-SQL Agent Training for Long-Horizon Tasks
By
–
MTSQL-R1 Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Training
-
FlashWorld: Fast High-Quality 3D Scene Generation Technology
By
–
FlashWorld
— AK (@_akhaliq) 16 octobre 2025
High-quality 3D Scene Generation within Seconds pic.twitter.com/jdwrdBuoHIFlashWorld High-quality 3D Scene Generation within Seconds
-
Attention Illuminates LLM Reasoning Through Preplan-and-Anchor Rhythm
By
–
Attention Illuminates LLM Reasoning The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization
-
Learning Symbolic World Models from Unguided Exploration
By
–
One Life to Learn
— AK (@_akhaliq) 15 octobre 2025
Inferring Symbolic World Models for Stochastic Environments from Unguided Exploration pic.twitter.com/Vcapsrcg1VOne Life to Learn Inferring Symbolic World Models for Stochastic Environments from Unguided Exploration
-
Mobile link for Akhaliq Veo3 1 Fast space
By
–
on mobile use this link: https://
akhaliq-veo3-1-fast.hf.space -
Error troubleshooting and mobile link solution
By
–
are you getting an error, also if your on mobile use this link: https://
akhaliq-veo3-1-fast.hf.space -
Apple DeepMMSearch-R1: Multimodal LLMs for Web Search
By
–
Apple presents DeepMMSearch-R1 Empowering Multimodal LLMs in Multimodal Web Search