Episodes (Page 12)
✨
GenEAva generates expressive cartoon avatars via diffusion.
✨
VCR-Bench evaluates video Chain-of-Thought reasoning for LVLMs.
✨
HalluMeasure automates hallucination detection in LLM outputs.
✨
Meta AI releases Llama 4 family of multimodal models.
✨
DISCIPL uses LMs to generate inference programs for self-steering.
✨
Amazon Q Developer in SageMaker Canvas uses natural language for ML.
✨
Adaptive SVD preserves knowledge in LLMs during continual learning.
✨
Meta AI released Llama 4 Scout, Maverick, and Behemoth, featuring multimodal capabilities.
✨
UniOcc is a new benchmark for 3D occupancy prediction in autonomous driving.
✨
A new method generates counterfactual explanations for graph-based predictions.
✨
Research addresses continual forgetting in pre-trained vision models.
✨
Optimizing adaptive OFDM for autonomous vehicles using millimeter wave.
✨
Improving video generation by aligning with human preferences.
✨
AnimeGamer enables infinite anime life simulation games.
✨
NoProp trains neural networks without backpropagation or forward/backward passes.
✨
ACPBench Hard evaluates LLM reasoning for automated planning.
✨
Episode: Efficient Training of Large Language Models
✨
Introduces Uni4D for dynamic 4D modeling from casual videos.
✨
Presents KDTalker for audio-driven talking portraits.
✨
Announces OLMo 2, fully open 7B and 13B language models.