activity
20162026
most citedLarge Language Models Understand and Can be Enhanced by Emotional Stimuli

61 citations · 201 across the 37 of their papers we have counts for

collaborators
Showing 2026Show all

6 papers · 1 filter

cs.AI2026

Preference Data Selection for Mitigating the Alignment Tax in Large Language Models

Minsu Kim, Jianxun Lian, Xing Xie +1

Aligning large language models to human preferences is crucial for real-world deployment but frequently incurs an alignment tax, leading to the catastrophic forgetting of pre-train…

cs.AI2026

Does Theory of Mind Improvement Really Benefit Human-AI Interactions? Empirical Findings from Interactive Evaluations

Nanxu Gong, Zixin Chen, Haotian Li +5

Improving the Theory of Mind (ToM) capability of Large Language Models (LLMs) is crucial for effective social interactions between these AI models and humans. However, the existing…

cs.AI2026

To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Tasks

Nanxu Gong, Haotian Li, Sixun Dong +3

Theory of Mind (ToM) assesses whether models can infer hidden mental states such as beliefs, desires, and intentions, which is essential for natural social interaction. Although re…

cs.CV2026

Proact-VL: A Proactive VideoLLM for Real-Time AI Companions

Weicai Yan, Yuhong Dai, Qi Ran +6

Proactive and real-time interactive experiences are essential for human-like AI companions, yet face three key challenges: (1) achieving low-latency inference under continuous stre…

cs.CL2026★ 1 cited

HumanLLM: Towards Personalized Understanding and Simulation of Human Nature

Yuxuan Lei, Tianfu Wang, Jianxun Lian +3

Motivated by the remarkable progress of large language models (LLMs) in objective tasks like mathematics and coding, there is growing interest in their potential to simulate human…

cs.IR2026

Why not Collaborative Filtering in Dual View? Bridging Sparse and Dense Models

Hanze Guo, Jianxun Lian, Xiao Zhou

Collaborative Filtering (CF) remains the cornerstone of modern recommender systems, with dense embedding--based methods dominating current practice. However, these approaches suffe…