Modality-Aware Expert Pruning for MoE-Based Multimodal Large Language Models
Paper • 2611.05927 • Published • 2 •
2
Librarian Bot (Bot)
librarian-bot
AI & ML interests
I am a friendly librarian bot working to help improve metadata on the 🤗 hub. Run by Hugging Face'ss Machine Learning Librarian (HF username: davanstrien)
Recent Activity
commentedon a paper about 5 hours ago
Modality-Aware Expert Pruning for MoE-Based Multimodal Large Language Models updated a dataset about 5 hours ago
librarian-bots/paper-recommendations-v2 commentedon a paper about 5 hours ago
PhyRubric: Rubric-Grounded Diagnostic Evaluation of Physical Commonsense in Text-to-Video GenerationOrganizations
commented a
paper about 5 hours ago
updated a
dataset about 5 hours ago
commented 18
papers about 5 hours ago
PhyRubric: Rubric-Grounded Diagnostic Evaluation of Physical Commonsense in Text-to-Video Generation
Paper • 2611.13576 • Published • 10 •
5
TorchCurves: Differentiable Parametric Curves
Paper • 2611.14516 • Published • 3 •
2
Tree-Ising: Repairing Language Models with Bit-Level Ising Edits
Paper • 2611.14210 • Published • 78 •
2
The Surprising Effectiveness of Deleting Weights in LLM Post-Training
Paper • 2611.10866 • Published • 16 •
3
Market-Conditioned JEPA: Learning from Unpredictable Market Shocks
Paper • 2611.13572 • Published • 5 •
3
Don't Just Trust Your Memory: Evidence-Linked Memory Reduces Memory-Induced Degradation in LLM Agents
Paper • 2611.14320 • Published • 60 •
7
Shortcuts and Boundaries: Verifiable RL Misalignment and Contextual Safety Learning
Paper • 2611.12865 • Published • 5 •
4
Are Causal Neuron Circuits Unique and Transferable?
Paper • 2611.13885 • Published • 6 •
5
Few-Step Streaming Video Generation via Condition Shifting
Paper • 2611.14033 • Published • 79 •
2
Prompt2Echo (P2Echo): Text-Prompted Multi-View Echocardiography Segmentation with One Model
Paper • 2611.13425 • Published • 1 •
2
DepthGraft: Structural Regularization through Hierarchical Cross-Layer KV Reconstruction
Paper • 2611.03142 • Published • 14 •
2
PhysOPSD: Code as Privilege for Physically Plausible Video Generation
Paper • 2611.12411 • Published • 8 •
3
Blind Variation, Selective Retention: Improving Creative Idea Generation in AI Scientists
Paper • 2611.12184 • Published • 4 •
3
ProTEUS: A mutation-resolved world model for closed-loop protein evolution
Paper • 2611.13348 • Published • 13 •
2
FINCOSMOS: An Open-World Evaluation Paradigm for Agents in Finance
Paper • 2611.07464 • Published • 22 •
4
EVAThink: Evidence Sufficiency-Guided Training for Adaptive Thinking in Video Understanding
Paper • 2611.13907 • Published • 3 •
3
World In-Context Learning for Agentic Robot Manipulation
Paper • 2611.12885 • Published • 67 •
3
Embodied First Exam
Paper • 2611.12177 • Published • 24 •
3