arXiv 论文速递

This paper examines the role of Novel View Synthesis (NVS) in geometric representation learning. In principle, NVS should reason about 3D scene structure, thereby enabling transfer…

首个面向动态场景逆图形学的代码生成基准，让智能体从视频重建可执行的4D场景，为评估具身与视觉推理能力提供了标准化测试平台。

We introduce 4DCodeBench, a benchmark for 4D inverse graphics through code generation, in which agents reconstruct dynamic scenes from video as executable graphics programs. To acc…

挑战持续学习中'遗忘即失败'的惯例，提出分层保留机制区分应遗忘与应保留信息，对世界模型的持续适应具有重要指导意义。

Continual learning treats degradation on previously seen data as evidence of failure, a convention inherited from settings with a stationary prediction target, where a correct labe…

Ribonucleic acid (RNA) functions through conformational changes that are not fully captured by static structures. However, large-scale standardized RNA dynamics data remain limited…

仅用单个立体相机和主动注视实现精细双臂操作，摆脱腕部相机依赖，为低成本、类人的机器人操作提供了新范式。

Inspired by human vision, we introduce a framework using active gaze to enable fine-grained bimanual manipulation with only a single stereo camera. EyeRobot 2.0 physically attends …

We study the minimization of sums of smooth strongly convex functions over undirected graphs, with each function held by one agent and communication restricted to neighbors in the …

提出用输出层梯度做后训练数据选择，大幅降低梯度计算开销，为LLM后训练数据筛选提供了更高效且可扩展的实用方案。

The choice of post-training data for large language models substantially affects downstream performance. Gradient-based data selection is a popular approach that ranks training dat…

Scarcity of labeled data limits development of deep learning biomarkers in oncology. We develop a two-stage AI model predicting pathological complete response (pCR) to neoadjuvant …

The dynamics of cells, organisms, and fluids are often modeled as probability distributions evolving over time. Reconstructing and extrapolating this evolution from unpaired snapsh…

LLM-guided evolutionary methods, such as AlphaEvolve, have emerged as powerful approaches for challenging computational optimization problems, such as circle packing. However, prio…

重新审视策略梯度中的探索与信用分配问题，提出以规划驱动学习，为强化学习及LLM后训练中的优化困境提供了新思路。

Policy-gradient methods are central to modern reinforcement learning, including LLM post-training. When they struggle, the usual suspects are exploration, credit assignment and act…

Masked diffusion language models (dLMs) offer a promising parallel alternative to autoregressive models for complex reasoning. However, they face a distinct credit-assignment chall…