8 papers
Dynamic Harness Search: Building Multi-Agent Systems Per-Query via Prediction
Som Sagar, Shasha Li, Hejie Cui +2
Agent harnesses specify the roles, instructions, tools, and communication structure used to solve a task, and the right harness depends on the query. Because the value of each desi…
RPTune: Learned Context Curation for LLM Catalog Search
Chuxuan Hu, Hejie Cui, Norman Huang +3
For small merchant businesses (SMBs) whose catalogs fit within a long-context LLM, full-catalog prompting offers a compelling alternative to multi-stage retrieval designed primaril…
AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation
Rishabh Agrawal, Hejie Cui, Shasha Li +2
A small trainable advisor can steer a frozen language-model executor using natural-language advice. In addition to learning from task rewards, the advisor can use feedback from com…
Customer-Agent: Overcoming Context Limitations in Ultra-Long Shopping Trajectories via Tool-Augmented Agents and RLVR
Hongye Liu, Rongmei Lin, Anurag Kashyap +4
Understanding customer shopping trajectories is essential for enabling personalized shopping experiences. However, shopping records (i.e., customer's search, clicks, purchases, etc…
TPO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning
Haixin Wang, Hejie Cui, Chenwei Zhang +7
Recent progress in multi-turn reinforcement learning (RL) has significantly improved reasoning LLMs' performances on complex interactive tasks. Despite advances in stabilization te…
Stepwise Penalization for Length-Efficient Chain-of-Thought Reasoning
Xintong Li, Sha Li, Rongmei Lin +10
Large reasoning models improve with more test-time computation, but often overthink, producing unnecessarily long chains-of-thought that raise cost without improving accuracy. Prio…
HeaPA: Difficulty-Aware Heap Sampling and On-Policy Query Augmentation for LLM Reinforcement Learning
Weiqi Wang, Xin Liu, Binxuan Huang +13
RLVR has become a standard recipe for training LLMs on reasoning tasks with verifiable outcomes, but when rollout generation dominates the cost, efficiency hinges on which prompts…
WebCoach: Self-Evolving Web Agents with Cross-Session Memory Guidance
Genglin Liu, Shijie Geng, Sha Li +4
Multimodal LLM-powered agents have recently demonstrated impressive capabilities in web navigation, enabling agents to complete complex browsing tasks across diverse domains. Howev…