DLite: History-free Speculative Decoding for Agentic AI
Paper • 2610.23787 • Published • 36
Owen Reed
owenreed
·
AI & ML interests
mostly healthcare AI, unit test generation, making models smaller, batching strategies, efficient inference, FlashAttention
Recent Activity
upvoted a paper 7 days ago
DLite: History-free Speculative Decoding for Agentic AI upvoted a paper 7 days ago
Lexical Reference Frames: Interpreting LM Residual Streams in Native SemanticsOrganizations
None yet
upvoted 2
papers 7 days ago
upvoted a paper 8 days ago
liked 2
datasets 9 days ago
upvoted 2
papers 9 days ago
upvoted 2
papers 22 days ago
upvoted 6
papers about 1 month ago
Harness-Zero: Harness Distillation via Agent-as-Harness
Paper • 2609.24974 • Published • 38
EvoOntology: A Self-Evolving Ontology Layer for Data Agents
Paper • 2609.15779 • Published • 131
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
Paper • 2609.19969 • Published • 237
RetireOPD: Self-Retiring On-Policy Distillation for Agentic Reinforcement Learning
Paper • 2609.20784 • Published • 58
Agora: Git as Shared Memory for Collective AutoResearch
Paper • 2609.18094 • Published • 59
NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction
Paper • 2609.10715 • Published • 205
liked a
model about 1 month ago
upvoted 4
papers about 1 month ago
ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement
Paper • 2609.14857 • Published • 271
Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction
Paper • 2609.13285 • Published • 83
How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus
Paper • 2609.15504 • Published • 37
Atria Dawn: The Dawn of Agentic Superintelligence
Paper • 2609.15818 • Published • 376