Article
YODAS v3: A 1 Million Hour Dataset for the Next Generation of Open Voice AI Research
espnet
• • 35
🤝 Open to Collab
Urro
urroxyz
P(doom) Rejects the premise
AI & ML interests
computational linguistics major 🤖🔎🔠
i am autistic. if i come off rude, i probably didn't mean to. please feel free to ask me for clarification.
Recent Activity
liked a dataset about 12 hours ago
espnet/yodas3 commentedon an article about 12 hours ago
YODAS v3: A 1 Million Hour Dataset for the Next Generation of Open Voice AI Research upvoted an article about 12 hours ago
YODAS v3: A 1 Million Hour Dataset for the Next Generation of Open Voice AI ResearchOrganizations
upvoted an article about 12 hours ago
upvoted a changelog about 15 hours ago
Hugging Face Changelog
P(doom) on user profiles
• 326
upvoted 8
papers about 21 hours ago
Minimal Witness Reinforcement Learning
Paper • 2610.07226 • Published • 19
EngramEdit: Decoupled Knowledge Updates in LLMs through Conditional Memory
Paper • 2610.10533 • Published • 7
STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization
Paper • 2609.38169 • Published • 109
Recurrent Looped Transformer
Paper • 2610.07591 • Published • 27
Gains and Collapse in On-Policy Distillation:A Reinforcement Learning Perspective
Paper • 2610.03185 • Published • 29
ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation
Paper • 2609.39306 • Published • 33
Questioning the Questions: Sustaining Self-Evolution in Reasoning Models
Paper • 2610.04299 • Published • 68
Semifactual Credit-Augmented Policy Optimization
Paper • 2609.40360 • Published • 38
upvoted a paper about 23 hours ago
upvoted 9
papers 1 day ago
Sharpen Without Search: On-Policy Distillation of Sequence-Level Power Distribution
Paper • 2610.06804 • Published • 6
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches
Paper • 2610.06647 • Published • 121
HuatuoGPT-3: RL-Only Domain Adaptation from Base Models
Paper • 2610.05966 • Published • 33
Labels Override Definitions in Jev-Style Typed Decision Models
Paper • 2610.02586 • Published • 7
Stepped MoE: Segment-Level Routing with Configurable Inference Complexity
Paper • 2610.07348 • Published • 4
OPD Before RL: Warm-Starting Rubric-Based RL with On-Policy Distillation
Paper • 2610.02781 • Published • 12
Multilinguality in Hybrid Attention LLMs
Paper • 2609.35378 • Published • 6
UNREAL: Unifying Retrieval and Long-Context with a Single Model
Paper • 2610.08463 • Published • 23
HLA: Expressive Hybrid Linear Attention via Chunk-Wise Dynamic Mixing
Paper • 2610.05842 • Published • 7