PerceptionDLM
— AK (@_akhaliq) 22 juin 2026
Parallel Region Perception with Multimodal Diffusion Language Models pic.twitter.com/0vZdGaAPoy
PerceptionDLM Parallel Region Perception with Multimodal Diffusion Language Models
@_akhaliq
-
PerceptionDLM: Parallel Region Perception with Multimodal Language Models
By
–
-
S-Agent: Spatial Tools Evoke Reasoning for Spatial Intelligence
By
–
S-Agent
— AK (@_akhaliq) 20 juin 2026
Spatial Tool-Use Elicits Reasoning for Spatial Intelligence pic.twitter.com/sbFDXYe1ZXS-Agent The Use of Spatial Tools Evokes Reasoning for Spatial Intelligence
-
GLM-5.2 free on Hugging Face via multiple inference providers for 6 hours
By
–
GLM-5.2 is free on Hugging Face Inference Providers via Zai, Together AI, Novita, Fireworks, DeepInfra for the next 6 hours. Configure it with Pi, opencode, Codex, Claude Code or any coding agent.
-
Decentralized Multi-Embodiment Collaboration using One VLA Policy
By
–
CHORUS
— AK (@_akhaliq) 12 juin 2026
Decentralized Multi-Embodiment Collaboration with One VLA Policy pic.twitter.com/QIyWXsMdiLCHORUS Decentralized Multi-Embodiment Collaboration with One VLA Policy
-
Standardization of inter-paradigm evaluation for tabular encoders
By
–
TRL-Bench Standardization of representation-level inter-paradigm evaluation for tabular encoders
-
Toward Generalist Autonomous Research via Hypothesis-Tree Refinement
By
–
Toward Generalist Autonomous Research via Hypothesis-Tree Refinement pic.twitter.com/NHDKezxAoY
— AK (@_akhaliq) 11 juin 2026Toward Generalist Autonomous Research via Hypothesis-Tree Refinement
-
Redefining MoE routers by power iteration on manifolds
By
–
Redefining Mixture-of-Experts routers with power iteration on manifolds
-
SCAIL-2: End-to-end character animation unification with context conditioning
By
–
SCAIL-2
— AK (@_akhaliq) 10 juin 2026
Unifying Controlled Character Animation with End-to-end In-Context Conditioning pic.twitter.com/bD0lUcHipJSCAIL-2 Unification of controlled character animation with end-to-end context conditioning
-
ABot-Earth 0.5: Generative 3D Model of the Earth
By
–
—
ABot-Earth 0.5 Generative 3D Model of the Earth
— -
Latent Spatial Memory for Video World Models
By
–
Latent Spatial Memory for Video World Models pic.twitter.com/sJIpofmrmQ
— AK (@_akhaliq) 9 juin 2026Latent Spatial Memory for World Models