18 papers
One Recipe, Many Harnesses: What Self-Evolution Encodes Across Languages and Models
Siqi Yang, Qianlan Yang, Yu-Xiong Wang +2
Self-evolving harnesses are closed-loop systems in which an agent inspects its own rollouts and edits its prompts, tools, and memory. They reliably improve coding agents in evaluat…
Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence
Xuanle Zhao, Qiushi Sun, Jingyu Xiao +16
While Large Language Models (LLMs) have substantially advanced text-to-code synthesis, many real programming tasks specify intent through visual artifacts such as screenshots, char…
Path-Coupled Bellman Flows for Distributional Reinforcement Learning
Boyang Xu, Qing Zou, Siqin Yang +1
Distributional reinforcement learning (DRL) models the full return distribution, but existing finite-support or quantile-based methods rely on projections, while recent flow-based…
Flexible Entropy Control in RLVR with a Gradient-Preserving Perspective
Kun Chen, Peng Shi, Fanfan Liu +4
Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a critical method for enhancing the reasoning capabilities of Large Language Models (LLMs). However, continuous…
Vision-and-Language Navigation for UAVs: Progress, Challenges, and a Research Roadmap
Hanxuan Chen, Jie Zheng, Siqi Yang +9
Vision-and-Language Navigation for Unmanned Aerial Vehicles (UAV-VLN) represents a pivotal challenge in embodied artificial intelligence, focused on enabling UAVs to interpret high…
Latent Wasserstein Adversarial Imitation Learning
Siqi Yang, Kai Yan, Alexander G. Schwing +1
Imitation Learning (IL) enables agents to mimic expert behavior by learning from demonstrations. However, traditional IL methods require large amounts of medium-to-high-quality dem…