Spatial reasoning in Vision-Language Models (VLMs) has increasingly benefited from tool-integrated pipelines that ground computation in explicit geometry. However, existing systems often compile queries into relatively open-ended artifacts, such as code, task constraints, tool plans, or gathered evidence, leaving what...
Xi-Wei Liu, Yu-Long Li, Xin-Lin Zhuang et al.· 0 citations
Large language models store substantial factual knowledge in their parameters, yet this knowledge is difficult to control, update, and reliably elicit at inference time. Existing memory-augmented approaches typically rely on a single auxiliary mechanism, such as retrieval, parameter updates, or neural memory, and there...
Yifan Lu, Ru-Bin Wei, Zi-Yun Zou et al.· 0 citations
Looped language models (LoopLMs) increase computational depth through parameter sharing, offering a path to scale inference computation without adding parameters. However, it remains unclear when additional recurrence is beneficial and how architectural choices affect its effectiveness. Through controlled experiments,...
Operator-level Adaptive Muon Orthogonalization (AMO), an observe-then-commit method that measures weight geometry by operator type early in training and then uses these signals to allocate the NS budget for the remainder of training, delivers consistent improvements over uniform-schedule Muon.
Xin-Lin Zhuang, Panyi Ouyang, Yicheng Li et al.· 0 citations
This work investigates two representative model families, LLaVA-1.5 and Qwen2.5, and provides a token-, layer-, and head-level account of how VLMs transform object grounding into spatial relations, showing that knowing where objects are is not equivalent to knowing how they relate.
Xi-Wei Liu, Yu-Long Li, Xin-Lin Zhuang et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.