UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation
Paper • 2611.16517 • Published • 12
GuoLiangTang
Tommy930
AI & ML interests
LLM，NLP，ML
Recent Activity
upvoted a paper about 7 hours ago
UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation upvoted a paper about 7 hours ago
G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation upvoted a paper about 7 hours ago
Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon AgentsOrganizations
None yet
upvoted 5
papers about 7 hours ago
G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation
Paper • 2611.13004 • Published • 39
Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents
Paper • 2611.17167 • Published • 26
On the Design of Hyper-Connections: Simplification, Stability, and Scalability
Paper • 2611.05984 • Published • 25
Duo: A Reciprocal World Model for Full-Duplex Interaction
Paper • 2611.15131 • Published • 33
upvoted 3
papers 1 day ago
Evolving Infra: Agent-Adaptive LLM Serving at Scale
Paper • 2611.12304 • Published • 13
GTAGym: AAA Game Tasks Environment for Evaluation and Reinforcement Learning
Paper • 2611.13028 • Published • 13
Incentivizing Self-Learning in Large Language Models via Reinforcement Learning
Paper • 2610.39671 • Published • 30
upvoted 7
papers 2 days ago
Blind Variation, Selective Retention: Improving Creative Idea Generation in AI Scientists
Paper • 2611.12184 • Published • 4
Tree-Ising: Repairing Language Models with Bit-Level Ising Edits
Paper • 2611.14210 • Published • 67
DepthGraft: Structural Regularization through Hierarchical Cross-Layer KV Reconstruction
Paper • 2611.03142 • Published • 16
PhysOPSD: Code as Privilege for Physically Plausible Video Generation
Paper • 2611.12411 • Published • 9
dNaViT: Discrete Codes are General-Purpose Visual Representations
Paper • 2611.12426 • Published • 36
e-means: outperforming k-means (in one-tenth the epochs)
Paper • 2611.09501 • Published • 11
SkinMind: Equipping Agentic RAG with Human-Inspired Graph Probing for Reliable Dermatology AI
Paper • 2611.08935 • Published • 4
upvoted 5
papers 3 days ago
FlyHash for LLM Agent Memory: An Empirical Validation and a New Application
Paper • 2611.05791 • Published • 5
InstantHMR : Real-Time Human Mesh Recovery on Your Phone
Paper • 2611.09195 • Published • 2
InfraRSI: Harnessing Recursive Self-Improvement for Training Infrastructure
Paper • 2611.11558 • Published • 34
Flide: Accelerating Decision Inference with Exact State Forking
Paper • 2610.42148 • Published • 37
WorldFoundry: Towards Unified Infrastructure for World Intelligence
Paper • 2611.10771 • Published • 59