4 papers
Gradient Starvation in Binary-Reward GRPO: Why Group-Mean Centering Fails and Why the Simplest Fix Works
Wenhua Nie, Jianan Wu, Junlin Liu +6
Group Relative Policy Optimization (GRPO) is a standard algorithm for reinforcement learning from verifiable rewards, but its group-mean-centered advantage can fail under binary re…
The Coupling Tax: How Shared Token Budgets Undermine Visible Chain-of-Thought Under Fixed Output Limits
Wenhua Nie, Junlin Liu, Jianan Wu +5
Chain-of-thought reasoning is often treated as a monotone way to improve language-model accuracy by letting a model think longer. We identify a countervailing effect, the coupling…
Reputation-Based Leader Election under Partial Synchrony: Towards a Protocol-Independent Abstraction with Enhanced Guarantees
Xuyang Liu, Zijian Zhang, Zhen Li +3
Leader election serves a well-defined role in leader-based Byzantine Fault Tolerant (BFT) protocols. Existing reputation-based leader election frameworks for partially synchronous…
Multi-source Multi-level Multi-token Ethereum Dataset and Benchmark Platform
Haoyuan Li, Mengxiao Zhang, Maoyuan Li +5
This paper introduces 3MEthTaskforce (https://3meth.github.io), a multi-source, multi-level, and multi-token Ethereum dataset addressing the limitations of single-source datasets.…