arXiv 论文速递

质疑文本到图像系统依赖互联网规模语言预训练的默认假设，探索无语言预训练的条件生成路径，对理解多模态模型的能力来源与训练成本有基础性意义。

Modern text-to-image (T2I) systems use internet-scale language pretraining for text-conditioned image generation. We ask whether language pretraining is necessary for strong T2I ge…

Multimodal large language models (MLLMs) have made rapid progress on visual reasoning, yet many tasks require evidence that is difficult to obtain through internal reasoning alone:…

解决多模态RLVR仅有任务级奖励、缺乏定位监督的痛点，用可视觉归因的在线自蒸馏提供细粒度指导，提升多模态强化学习的可解释性与效果。

Multimodal reinforcement learning with verifiable rewards (RLVR) provides reliable task-level supervision, yet offers limited guidance for localizing errors in visual reasoning. On…

Reinforcement learning enables mobile robots to make fast and scalable navigation decisions in constrained human-shared environments. However, directly extending single-robot appro…

系统研究物理递归自我改进的机制与缩放规律，提出观测性缩放律，为LLM作为策略在具身场景中的能力边界与泛化提供可量化框架。

In this paper, we systematically investigate the mechanisms, scaling behavior, and generalization of physical Recursive self-improvement. We introduce PhysRSI-eval to quantify phys…

Long-horizon agents must learn heterogeneous capabilities from interleaved trajectories. Updating shared parameters to learn new capabilities can degrade previously acquired ones, …

A seemingly reasonable push in Sokoban can make the puzzle unsolvable. To avoid such mistakes, an agent has to anticipate the consequences of its actions before execution, especial…

针对世界模型只能拟合专家轨迹、难以泛化到任意目标的问题，研究超越专家示范的动力学泛化，对基于模型的规划与想象式智能体至关重要。

The goal of a world model is to capture the dynamics of an environment. This has the great appeal that it allows agents to plan and simulate a course of actions with the help of a …

Systems that combine several large language models (LLMs) often use one context for many questions. When the context is passed between models as text, the receiving model processes…

Bilevel optimization provides a framework for machine learning tasks such as hyperparameter optimization, data reweighting, and meta-learning. Existing Newton-based bilevel methods…

LLM agents that compute the right answer can still answer wrongly. Agent harnesses increasingly keep long tool outputs out of the context window, saving them to files that the agen…

提出基于Kronecker Delta规则的线性注意力更新机制KronLSTM，改进关联矩阵记忆的长上下文建模，是架构层面兼具理论动机与实用价值的贡献。

We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM family with …