2 citations · 2 across the 9 of their papers we have counts for
7 papers · 1 filter
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
DeepSeek-AI, :, Anyi Xu +585
The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation…
EmbodiedMind: Adaptive Data Curation and Prefix-Tree Reinforcement Learning for Efficient Embodied Intelligence
Feifan Wang, Zongbing Zhang, Yu Zhang +8
Training embodied foundation models typically requires massive-scale datasets and extensive computational resources, yet often suffers from three critical limitations: (1) ineffici…
CA-OPD: Confidence-Aware On-Policy Distillation for Structured Visual Prediction
Menghao Li, Linjie Mu, Yin Wang +6
Autoregressive vision language models unify heterogeneous perception tasks but are highly susceptible to compounding errors. On-policy distillation (OPD) bridges the training-infer…
AgenticRS-EnsNAS: Ensemble-Decoupled Self-Evolving Architecture Search
Yun Chen, Moyu Zhang, Jinxin Hu +2
Neural Architecture Search (NAS) deployment in industrial production systems faces a fundamental validation bottleneck: verifying a single candidate architecture pi requires evalua…
OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
Zhuofeng Li, Dongfu Jiang, Xueguang Ma +7
Training deep research agents requires long-horizon trajectories that interleave search, evidence aggregation, and multi-step reasoning. However, existing data collection pipelines…
DSDR: Dual-Scale Diversity Regularization for Exploration in LLM Reasoning
Zhongwei Wan, Yun Shen, Zhihao Dou +9
Reinforcement learning with verifiers (RLVR) is a central paradigm for improving large language model (LLM) reasoning, yet existing methods often suffer from limited exploration. P…
Vid2Sid: Videos Can Help Close the Sim2Real Gap
Kevin Qiu, Yu Zhang, Marek Cygan +1
Calibrating a robot simulator's physics parameters (friction, damping, material stiffness) to match real hardware is often done by hand or with black-box optimizers that reduce err…