Agentic robots must turn a high-level goal into actions while keeping the physical conditions needed for future steps valid. A completed action does not ensure that its effects will persist: a later disturbance can invalidate support that the remaining plan still assumes. We present LOGIV, a graph engineering framework...
En-Qiao Lu, Yu-Xiang Xie, Xingrui Yu et al.· 0 citations
Multimodal intelligence can be viewed as human-guided co-evolution between worlds and agents through interaction. However, scaling this process faces two bottlenecks: constructing heterogeneous, verifiable environments is costly, and manual Harness iteration cannot keep pace with growing task diversity and complexity....
Wei-Dong Tang, Rui-Heng Wang, Jing-Xuan Zhang et al.· 2 citations
REA (Routed Embodied Agent) is proposed, a unified runtime where a deliberative System 2, an embodied router System 1 and reactive System 0 policies cooperate under a single golden rule - no running policy waits on the brain.
P. Zhou, Yi-Xin Ma, Zhengfeiyang Zhang et al.· 0 citations
The Trinity is introduced, a recursive self-improvement framework coupling world generation with Harness evolution through three roles: Creator, Explorer, and Evolver, which introduces Capability-Driven Real2Sim2Real: for heterogeneous tasks at the same node, Creator abstracts their common requirements into simplified,...
Wei-Dong Tang, Rui-Heng Wang, Jing-Xuan Zhang et al.· 0 citations
Online reinforcement learning (RL) fine-tunes diffusion models with black-box rewards, but reverse-process methods such as Flow-GRPO are limited in efficiency by poor credit assignment, while forward-process methods such as DiffusionNFT are efficient yet unstable without carefully tuned exponential-moving-average (EMA)...
Zheng-Lin Wan, Jia-Yu Chen, Zhengfeiyang Zhang et al.· 0 citations
An agent whose tool keeps returning nothing useful should stop relying on it. In a retrieval environment with controlled source failures, we separate how agents judge results from what they do. We compare stopping at the same step after longer and shorter runs of results the agent judged useless; this contrast is zero...
Chubin Zhang, Zheng-Lin Wan, Xingrui Yu et al.· 0 citations
Time-Annealed Perturbation Sampling (TAPS) is proposed, a training-free inference strategy that samples nearby conditioning trajectories through time-aware, manifold-constrained perturbations, and consistently improves semantic and lexical diversity across open-ended and instruction-following generation tasks.
Jing-Xuan Wu, Zheng-Lin Wan, Xing-Rui Yu et al.· 4 citations
CUA-Sandbox is introduced, which separates private state capsules from shared runtimes through state-scoped execution and transactional lifecycle operations, including resets and branches, while retaining the original software interfaces and task evaluators.
Xin Yan, Zheng-Bo Jiao, Jia-Qi Liu et al.· arXiv.org· 2 citations
VIsion Pruning Router is proposed, a lightweight VIsion Pruning Router that adaptively selects the pruning strategy predicted to be best suited to each input at a specified pruning level that proves effective across various MLLM backbones and yields consistent gains on unseen benchmarks, highlighting the potential of s...
Hai-Jin Liang, P. Zhou, Zheng-Lin Wan et al.· 0 citations
This work proposes Reinforcement Learning with Human-Engine Verification (RLHEV), a post-training paradigm that combines dense engine signals with implicit human acceptance feedback from the development process to support RL post-training.
P. Zhou, Hesong Wang, Zhengfeiyang Zhang et al.· 1 citation
Reinforcement Learning with Verifiable Rewards (RLVR) makes Multimodal Large Language Models more accurate, but the gains are brittle: simply paraphrasing a question or changing the prompt template can degrade them, which challenges reliable deployment in high-stakes scenarios like medical VQA. We trace this to two iss...
P. Zhou, Zhiwei Tang, Xiaopeng Peng et al.· 0 citations
Spiking neural networks (SNNs) offer a path to energy-efficient language modeling through sparse encoding and event-driven computation, but training capable spiking language models from scratch remains difficult. A practical alternative is ANN-to-SNN migration through knowledge distillation (KD), where a pretrained art...
En-Qiao Lu, Xingrui Yu, Yi-Wei Fu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.