Research publications and technical reports, listed in reverse chronological order.
This technical report introduces Kimi K3, an open-weight, native multimodal Mixture-of-Experts model with 2.8 trillion total parameters, 104 billion activated parameters, and a one-million-token context window for long-horizon coding, ag...
To accurately model the intricate nature of length bias and facilitate more effective bias mitigation, it proposes FiMi-RM (Bias Fitting to Mitigate Length Bias of Reward Model in RLHF), a framework that autonomously learns and corrects ...
It introduces a Response-conditioned Bradley-Terry (Rc-BT) model that enhances the model's capability in length bias mitigating and length instruction following, through training on the augmented dataset. Furthermore, it proposes the Rc-...
The rise of reasoning models necessitates large-scale verifiable data, for which programming tasks serve as an ideal source. To address this, we propose a Feedback-Driven Iterative Framework for comprehensive test case construction and r...
The research identifies a critical oversight in existing techniques, which predominantly focus on comparing responses while neglecting valuable latent signals embedded within prompt inputs, and which only focus on preference disparities ...
It is the first to systematically investigate the effectiveness and underlying mechanisms of activation engineering for mitigating hallucinations in VideoLLMs. And it proposes a temporal-aware activation engineering framework for VideoLL...
This paper proposes a Heterogeneous Network based on Contrastive Learning (HCLNet). HCLNet aims to learn high-level representation from unlabeled PolSAR data for few-shot classification according to multi-features.
This paper proposes the Flowmind2digital method and hdFlowmind dataset to address the convertion of hand-drawn flowchart/mindmap.