6 papers
New Terms, New Toxicity: Consensus-based Chinese Neologism Toxicity Detection via Search-Augmented LLMs
Shiyao Cui, QingLin Zhang, Di Wang +7
Neologisms, emerging terms in meaning or form, can serve as new vehicles for toxic expression, like "country girl" as a stigmatizing label targeting feminism. Such toxic neologisms…
Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards
Zhengzhao Ma, Xueru Wen, Boxi Cao +6
Reinforcement Learning from Verifiable Rewards (RLVR) significantly enhances large language models (LLMs) reasoning but severely suffers from calibration degeneration, where models…
P^2O: Joint Policy and Prompt Optimization
Xinyu Lu, Kaiqi Zhang, Jinglin Yang +6
Reinforcement Learning with Verifiable Rewards (RLVR) enhances Large Language Model (LLM) reasoning but suffers from advantage collapse on ``hard samples'' where all rollouts fail.…
Enhancing LLM-Based Agents via Global Planning and Hierarchical Execution
Junjie Chen, Haitao Li, Jingli Yang +2
Intelligent agent systems based on Large Language Models (LLMs) have shown great potential in real-world applications. However, existing agent frameworks still face critical limita…
PoAct: Policy and Action Dual-Control Agent for Generalized Applications
Guozhi Yuan, Youfeng Liu, Jingli Yang +6
Based on their superior comprehension and reasoning capabilities, Large Language Model (LLM) driven agent frameworks have achieved significant success in numerous complex reasoning…
LegalAgentBench: Evaluating LLM Agents in Legal Domain
Haitao Li, Junjie Chen, Jingli Yang +10
With the increasing intelligence and autonomy of LLM agents, their potential applications in the legal domain are becoming increasingly apparent. However, existing general-domain b…