Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding
Paper • 2604.26779 • Published • 13
M Saad Salman
MSS444
AI & ML interests
None yet
Recent Activity
upvoted a paper 2 days ago
Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding upvoted a paper 2 days ago
Large Language Models Explore by Latent DistillingOrganizations
None yet
upvoted 12
papers 2 days ago
Large Language Models Explore by Latent Distilling
Paper • 2604.24927 • Published • 72
Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization
Paper • 2604.24952 • Published • 6
Compliance versus Sensibility: On the Reasoning Controllability in Large Language Models
Paper • 2604.27251 • Published • 8
Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence
Paper • 2604.24954 • Published • 21
Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling
Paper • 2604.27039 • Published • 24
Intern-Atlas: A Methodological Evolution Graph as Research Infrastructure for AI Scientists
Paper • 2604.28158 • Published • 45
Efficient Training on Multiple Consumer GPUs with RoundPipe
Paper • 2604.27085 • Published • 38
Co-Evolving Policy Distillation
Paper • 2604.27083 • Published • 61
Heterogeneous Scientific Foundation Model Collaboration
Paper • 2604.27351 • Published • 206
From Skill Text to Skill Structure: The Scheduling-Structural-Logical Representation for Agent Skills
Paper • 2604.24026 • Published • 18
Let ViT Speak: Generative Language-Image Pre-training
Paper • 2605.00809 • Published • 26
upvoted 6
papers 10 days ago
Convergent Evolution: How Different Language Models Learn Similar Number Representations
Paper • 2604.20817 • Published • 7
Abstain-R1: Calibrated Abstention and Post-Refusal Clarification via Verifiable RL
Paper • 2604.17073 • Published • 9
Scaling Test-Time Compute for Agentic Coding
Paper • 2604.16529 • Published • 11
Expert Upcycling: Shifting the Compute-Efficient Frontier of Mixture-of-Experts
Paper • 2604.19835 • Published • 19
Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
Paper • 2604.13602 • Published • 32
Near-Future Policy Optimization
Paper • 2604.20733 • Published • 74
upvoted 2
papers 16 days ago