PPL
Chao Zhang
Papers (10)
EAGLE-2: Faster Inference of Language Models with Dynamic Draft TreesEAGLE-3: Scaling up Inference Acceleration of Large Language Models via Training-Time TestEAGLE: Speculative Sampling Requires Rethinking Feature UncertaintySame Evidence, Different Answers: Canonical-Context On-Policy Distillation for Multi-Turn Language ModelsOScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and BeyondRevisiting DAgger in the Era of LLM-AgentsEvolutionary Task Discovery: Advancing Reasoning Frontiers via Skill Composition and Complexity ScalingSpeech-based Psychological Crisis Assessment using LLMsExploration-Driven Optimization for Test-Time Large Language Model ReasoningBeyond Steering Vector: Flow-based Activation Steering for Inference-Time Intervention