11 papers
VidHarness: Evolving Agent Harnesses for Cost-Efficient Long Video Understanding
Susan Liang, Jianmin Wu, Daxiang Dong
Vision-language models (VLMs) can answer questions about hour-long videos, but processing every frame is prohibitively expensive, even though the evidence for a question usually sp…
WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games
Wenyu Zhang, Guoliang You, Tianlun +8
Coding agents are increasingly used as application builders, yet many evaluations still focus on source code, repository-level tests, or intermediate traces rather than the deliver…
Rollout Pass-Rate Control: Steering Binary-Reward RL Toward Its Most Informative Regime
Tianshu Zhu, Wenyu Zhang, Xiaoying Zuo +8
Agentic reinforcement learning (RL) for software engineering spends much of its compute on stateful trajectories whose grouped binary rewards are highly skewed and weakly contrasti…
AEM: Adaptive Entropy Modulation for Multi-Turn Agentic Reinforcement Learning
Haotian Zhao, Songlin Zhou, Yuxin Zhang +9
Reinforcement learning (RL) has substantially improved the ability of large language model (LLM) agents to interact with environments and solve multi-turn tasks. However, effective…
Qianfan-OCR: A Unified End-to-End Model for Document Intelligence
Daxiang Dong, Mingming Zheng, Dong Xu +17
We present Qianfan-OCR, a 4B-parameter end-to-end vision-language model that unifies document parsing, layout analysis, and document understanding within a single architecture. It…
Detached Skip-Links and -Probe: Decoupling Feature Aggregation from Gradient Propagation for MLLM OCR
Ziye Yuan, Ruchang Yao, Chengxin Zheng +3
Multimodal large language models (MLLMs) excel at high-level reasoning yet fail on OCR tasks where fine-grained visual details are compromised or misaligned. We identify an overloo…
SWE-Hub: A Unified Production System for Scalable, Executable Software Engineering Tasks
Yucheng Zeng, Shupeng Li, Daxiang Dong +11
Progress in software-engineering agents is increasingly constrained by the scarcity of executable, scalable, and realistic data for training and evaluation. This scarcity stems fro…
QianfanHuijin Technical Report: A Novel Multi-Stage Training Paradigm for Finance Industrial LLMs
Shupeng Li, Weipeng Lu, Linyun Liu +16
Domain-specific enhancement of Large Language Models (LLMs) within the financial context has long been a focal point of industrial application. While previous models such as Bloomb…
Qianfan-VL: Domain-Enhanced Universal Vision-Language Models
Daxiang Dong, Mingming Zheng, Dong Xu +32
We present Qianfan-VL, a series of multimodal large language models ranging from 3B to 70B parameters, achieving state-of-the-art performance through innovative domain enhancement…
Tree-of-Code: A Hybrid Approach for Robust Complex Task Planning and Execution
Ziyi Ni, Yifan Li, Daxiang Dong
The exceptional capabilities of large language models (LLMs) have substantially accelerated the rapid rise and widespread adoption of agents. Recent studies have demonstrated that…
Tree-of-Code: A Tree-Structured Exploring Framework for End-to-End Code Generation and Execution in Complex Task Handling
Ziyi Ni, Yifan Li, Ning Yang +3
Solving complex reasoning tasks is a key real-world application of agents. Thanks to the pretraining of Large Language Models (LLMs) on code data, recent approaches like CodeAct su…