13 papers
Do Rollups Self-Organize? Multi-Agent Reinforcement Learning in the EIP-4844 Blob Fee Market
Alessio Brini
EIP-4844 gave Ethereum rollups a separate fee dimension for data, governed by a multiplicative update rule that targets three ``blobs'' per block. Whether this target is reached de…
How to Loop MoE: Flatten the Experts, Untie the Attention
Shouren Wang, Chuang Ma, Mohsen Hariri +6
Looped Transformers reuse one block of layers several times: by spending extra computation they push a model of fixed size further, and so use its parameters more fully; while spar…
Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility
Mohsen Hariri, Weicong Chen, Nahal Shahini +11
Large language models can solve harder reasoning problems with more inference-time compute. The term "test-time scaling," however, covers several inference algorithms: extending de…
Agent Team Work Zone: An Automated, Persistent Workspace for Long-Lived Claude Code Agent Teams
Shouren Wang
Large Language Model (LLM) agents have significantly improved coding and programming workflows. Claude Code, in particular, is one of the most powerful LLM coding agents and is cap…
MemTrace: Probing What Final Accuracy Misses in Long-Term Memory
Xianxuan Long, Zhikai Chen, Shenglai Zeng +3
LLM agents increasingly maintain long-term memory of user facts across sessions. Yet such memory is usually evaluated by aggregating accuracy over question rows or episodes. Becaus…
CausalGuard: Conformal Inference under Graph Uncertainty
Vikash Singh, Weicong Chen, Debargha Ganguly +12
Estimating treatment effects from observational data requires choosing an adjustment set, but valid adjustment depends on an unknown causal graph. Graph misspecification can cause…
Reliability-Gated Source Anchoring for Continual Test-Time Adaptation
Vikash Singh, Debargha Ganguly, Weicong Chen +8
Continual test-time adaptation (CTTA) updates a pretrained model online on an unlabeled, non-stationary stream while anchoring it to a frozen source checkpoint. This anchor is usef…
AgentCE-Bench: Agent Configurable Evaluation with Scalable Horizons and Controllable Difficulty under Lightweight Environments
Wang Yang, Chaoda Song, Xinpeng Li +7
Existing Agent benchmarks suffer from two critical limitations: high environment interaction overhead (up to 41\% of total evaluation time) and imbalanced task horizon and difficul…
Path-Lock Expert: Separating Reasoning Mode in Hybrid Thinking via Architecture-Level Separation
Shouren Wang, Wang Yang, Chuang Ma +7
Hybrid-thinking language models expose explicit /think and /no_think modes, but current designs do not separate them cleanly. Even in /no_think mode, models often emit long and sel…
When Domains Interact: Asymmetric and Order-Sensitive Cross-Domain Effects in Reinforcement Learning for Reasoning
Wang Yang, Shouren Wang, Chaoda Song +6
Group Relative Policy Optimization (GRPO) has become a key technique for improving reasoning abilities in large language models, yet its behavior under different domain sequencing…
Mid-Think: Training-Free Intermediate-Budget Reasoning via Token-Level Triggers
Wang Yang, Debargha Ganguly, Xinpeng Li +5
Hybrid reasoning language models are commonly controlled through high-level Think/No-think instructions to regulate reasoning behavior, yet we found that such mode switching is lar…
Demystifying Hybrid Thinking: Can LLMs Truly Switch Between Think and No-Think?
Shouren Wang, Wang Yang, Xianxuan Long +4
Hybrid thinking enables LLMs to switch between reasoning and direct answering, offering a balance between efficiency and reasoning capability. Yet our experiments reveal that curre…
Enhancing Player Enjoyment with a Two-Tier DRL and LLM-Based Agent System for Fighting Games
Shouren Wang, Zehua Jiang, Fernando Sliva +2
Deep reinforcement learning (DRL) has effectively enhanced gameplay experiences and game design across various game genres. However, few studies on fighting game agents have focuse…