6 papers
RSMeM: Knowledge-Enhanced Memory Evolution for Remote Sensing Agents with Systematic Evaluation
Bingxian Wu, Yu Zhang, Zonghao Guo +9
Geoscience research requires complex analysis and domain expertise, with remote sensing (RS) observations as a key foundation. However, existing RS agents built on general-purpose…
Imagination Helps Visual Reasoning, But Not Yet in Latent Space
You Li, Chi Chen, Yanghao Li +4
Latent visual reasoning aims to mimic human's imagination process by meditating through hidden states of Multimodal Large Language Models. While recognized as a promising paradigm…
Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation
Yichen Zhang, Da Peng, Zonghao Guo +19
A recent cutting-edge topic in multimodal modeling is to unify visual comprehension and generation within a single model. However, the two tasks demand mismatched decoding regimes…
InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation
Weilin Zhao, Zihan Zhou, Zhou Su +10
Long-sequence processing is a critical capability for modern large language models. However, the self-attention mechanism in the standard Transformer architecture faces severe comp…
MiniCPM4: Ultra-Efficient LLMs on End Devices
MiniCPM Team, Chaojun Xiao, Yuxuan Li +80
This paper introduces MiniCPM4, a highly efficient large language model (LLM) designed explicitly for end-side devices. We achieve this efficiency through systematic innovation in…
Distance between Relevant Information Pieces Causes Bias in Long-Context LLMs
Runchu Tian, Yanghao Li, Yuepeng Fu +10
Positional bias in large language models (LLMs) hinders their ability to effectively process long inputs. A prominent example is the "lost in the middle" phenomenon, where LLMs str…