4 papers
Cultural Palette: Pluralising Culture Alignment via Multi-agent Palette
Jiahao Yuan, Zixiang Di, Shangzixin Zhao +4
Large language models (LLMs) face challenges in aligning with diverse cultural values despite their remarkable performance in generation, which stems from inherent monocultural bia…
ReflectDiffu:Reflect between Emotion-intent Contagion and Mimicry for Empathetic Response Generation via a RL-Diffusion Framework
Jiahao Yuan, Zixiang Di, Zhiqing Cui +2
Empathetic response generation necessitates the integration of emotional and intentional dynamics to foster meaningful interactions. Existing research either neglects the intricate…
Reversal of Thought: Enhancing Large Language Models with Preference-Guided Reverse Reasoning Warm-up
Jiahao Yuan, Dehui Du, Hao Zhang +2
Large language models (LLMs) have shown remarkable performance in reasoning tasks but face limitations in mathematical and complex logical reasoning. Existing methods to improve LL…
LLMSR@XLLM25: Less is More: Enhancing Structured Multi-Agent Reasoning via Quality-Guided Distillation
Jiahao Yuan, Xingzhe Sun, Xing Yu +4
The LLMSR@XLLM25 formulates a low-resource structural reasoning task that challenges LLMs to generate interpretable, step-by-step rationales with minimal labeled data. We present L…