Recent focus in agentic reinforcement learning (RL) seeks to improve credit assignment in long-horizon tasks. One promising direction combines reinforcement learning with verifiable rewards (RLVR) and on-policy self-distillation (OPSD), integrating trajectory-level advantages with dense token-level supervision. This li...
Fan Zhang, Hao Wu, Jin-Yang Wu et al.· 0 citations
Instruction-guided video editing has made significant progress, yet existing methods use multimodal large language models (MLLMs) primarily as semantic encoders, so they often fall short in working with implicit edits that require causal or semantic reasoning. To bridge this fundamental gap in video editing, we propose...
Dong-Hao Zhou, Hao-Yang He, Fan Zhang et al.· 0 citations
The Noisy Test-time Reinforcement Learning framework (NTRL-Code) is proposed, which enables robust self-evolution of code LLMs using only unlabeled noisy data during the testing stage, and employs an abstract-syntax-tree (AST)-based structural aggregation mechanism to estimate a proxy target from multiple candidate pro...
Xi-Kai Yang, Hieu Trung Nguyen, Dun-Yuan Xu et al.· 0 citations
An injectively aligned leave-one-cell-out protocol over support x operation grids and SO-OPF, a readout that factors cell energy into support salience and a competitive operation posterior, and a readout that factors cell energy into support salience and a competitive operation posterior are introduced.
Zhong-Yao Wang, Wan-Li Ouyang, Tao-Yong Cui et al.· 0 citations
JEPA-Anything, a domain-agnostic framework based on orthogonal predictive factorization (OPF), achieves the lowest one-step and 100-step molecular errors among compared methods in all four systems and supports a common factorized predictive principle across heterogeneous worlds.
Tao-Yong Cui, Zhong-Yao Wang, Xin-Yue Xu et al.· 2 citations
This work introduces VideoGAIA, an agentic video understanding benchmark for general artificial intelligence (AI) assistants, and formulates video understanding as a multi-turn, tool-augmented interaction process, where models must iteratively perceive videos, invoke external tools, gather complementary information, an...
Fan Zhang, Guang-Ming Yao, Jin-Yang Wu et al.· 1 citation
This work introduces a real-world multi-modal benchmark for adverse-weather autonomous driving, designed with three capability dimensions: observability awareness, spatial reliability, and risk-aware decision-making, enabling fine-grained diagnosis of model behavior under degraded observations.
VideoCoCo, an agentic dual-engine framework in which executable Blender code serves as a process-level chain of thought, demonstrates that executable code provides an effective, controllable, and inspectable intermediate representation for physically consistent video generation.
Haodong Li, Tianfei Ren, Xiaoxiao Ma et al.· arXiv.org· 8 citations
Method, a latent world-modeling framework based on orthogonal predictive factorization, is introduced, a latent world-modeling framework based on orthogonal predictive factorization that can be used by a readout, decoder, planner, or autoregressive rollout of an underlying system.
