15 papers
: Twin-Critic Training for Token-Level Thoughts in Reinforcement Mid-Training
Nan Qiao, Yebin Yang, Weinong Wang +10
Reinforcement mid-training lets language models learn internal thoughts from unlabeled text, but efficient token-level credit assignment remains challenging. Existing group-relativ…
Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs
Xinming Wang, Weinong Wang, Hongming Yang +13
Hybrid-thinking multimodal large language models (MLLMs) allow a single model to alternate between deliberative thinking and latency-efficient non-thinking inference. Although thes…
HunyuanOCR-1.5: Making Lightweight OCR VLMs Faster and Better
Gengluo Li, Xingyu Wan, Shangpin Peng +20
We present HunyuanOCR-1.5, a lightweight end-to-end OCR-specialized vision-language model. HunyuanOCR unifies document parsing, text spotting, information extraction, text-image tr…
HyMobileAgent: Data-Environment Co-Scaling for Efficient GUI Agents
Hy Vision Team, Huawen Shen, Zhengyang Tang +20
As large multimodal models move from understanding content to operating on digital environments, mobile GUI has emerged as a challenging and consequential testbed for digital embod…
ChartArena: Benchmarking Chart Parsing across Languages, Scenarios, and Formats
Shangpin Peng, Gengluo Li, Xingyu Wan +10
Charts are a primary medium for conveying quantitative and relational information, yet systematically evaluating chart parsing models remains difficult. Existing benchmarks focus o…
StrucTab: A Structured Optimization Framework for Table Parsing
Gengluo Li, Shangpin Peng, Chengquan Zhang +10
Table parsing aims to convert table images into structured, machine-readable representations, a task requiring the joint perception of complex spatial layouts and textual content.…
PhoneBuddy: Training Open Models for Agentic Phone Use
Zhengyang Tang, Xin Lai, Pengyuan Lyu +23
Phones are becoming an important execution surface for general-purpose agents, but training open models for reliable phone use remains difficult because the environment that matter…
PhoneHarness: Harnessing Phone-Use Agents through Mixed GUI, CLI, and Tool Actions
Chenxin Li, Zhengyao Fang, Zhengyang Tang +18
Phone agents are increasingly expected to complete real mobile workflows rather than merely predict the next screen action. However, much of the current mobile-agent literature sti…
Chronicles-OCR: A Cross-Temporal Perception Benchmark for the Evolutionary Trajectory of Chinese Characters
Gengluo Li, Shangpin Peng, Xingyu Wan +16
Vision Large Language Models (VLLMs) have achieved remarkable success in modern text-rich visual understanding. However, their perceptual robustness in the face of the continuous m…
Safe, or Simply Incapable? Rethinking Safety Evaluation for Phone-Use Agents
Zhengyang Tang, Yi Zhang, Chenxin Li +18
When a phone-use agent avoids harm, does that show safety, or simply inability to act? Existing evaluations often cannot tell. A harmful outcome may be avoided because the agent re…
PhoneWorld: From Real-App Trajectories to Dynamic and Verifiable Environments for Phone-Use Agents
Yuxuan Liu, Xin Lai, Junyi Li +21
Real applications provide the training setting closest to phone-agent deployment, but are difficult to reset, scale safely, and verify programmatically. Static screenshots and inte…
Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe
Wenjin Hou, Shangpin Peng, Weinong Wang +13
On-policy distillation (OPD) has recently emerged as an effective post-training paradigm for consolidating the capabilities of specialized expert models into a single student model…
HunyuanOCR Technical Report
Hunyuan Vision Team, Pengyuan Lyu, Xingyu Wan +29
This paper presents HunyuanOCR, a commercial-grade, open-source, and lightweight (1B parameters) Vision-Language Model (VLM) dedicated to OCR tasks. The architecture comprises a Na…
Mitigating Object Hallucinations via Sentence-Level Early Intervention
Shangpin Peng, Senqiao Yang, Li Jiang +1
Multimodal large language models (MLLMs) have revolutionized cross-modal understanding but continue to struggle with hallucinations - fabricated content contradicting visual inputs…
Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs
Shangpin Peng, Weinong Wang, Zhuotao Tian +7
Direct Preference Optimization (DPO) has emerged as a cornerstone of reinforcement learning from human feedback (RLHF) due to its simplicity and efficiency. However, existing DPO-b…