Related papers: CoVe: Training Interactive Tool-Use Agents via Con…

Multimodal large language models increasingly solve vision-centric tasks by calling external tools for visual inspection, OCR, retrieval, calculation, and multi-step reasoning. Current tool-using agents usually expose the executed tool…

Autonomous Driving (AD) systems have made notable progress, but their performance in long-tail, safety-critical scenarios remains limited. These rare cases contribute a disproportionate number of accidents. Vision-Language Action (VLA)…

Reinforcement learning (RL) has become a prevalent paradigm for training tool calling agents, which typically requires online interactive environments. Existing approaches either rely on training data with ground truth annotations or…

Large language models like GPT-4 are resource-intensive, but recent advancements suggest that smaller, specialized experts can outperform the monolithic models on specific tasks. The Collaboration-of-Experts (CoE) approach integrates…

This study investigates the potential of virtual reality (VR) for enhancing sales skills training using a Cave Automatic Virtual Environment (CAVE). VR technology enables users to practice interpersonal and negotiation skills in controlled,…

Designing effective algorithmic components remains a fundamental obstacle in tackling NP-hard combinatorial optimization problems (COPs), where solvers often rely on carefully hand-crafted strategies. Despite recent advances in using large…

Research on applications of reinforcement learning (RL) to large language models has mostly been focused on single-turn problems, such as mathematical reasoning or single-shot code generation. While these problems can be viewed as…

We humans can impeccably search for a target object, given its name only, even in an unseen environment. We argue that this ability is largely due to three main reasons: the incorporation of prior knowledge (or experience), the adaptation…

Large-scale, high-quality interaction trajectories are essential for advancing mobile Graphical User Interface (GUI) agents. While existing methods typically rely on labor-intensive human demonstrations or automated model exploration to…

Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of LLMs, yet a fundamental limitation remains: models cannot learn from problems that are too difficult to solve under their current policy, as these…

Single object tracking (SOT) research falls into a cycle -- trackers perform well on most benchmarks but quickly fail in challenging scenarios, causing researchers to doubt the insufficient data content and take more effort to construct…

Recent advances in large language models (LLMs) have enabled a new generation of autonomous agents that operate over sustained periods and manage sensitive resources on behalf of users. Trusted for their ability to act without direct…

As AI agents take on increasingly long-running tasks involving sophisticated planning and execution, there is a corresponding need for novel interaction designs that enable deeper human-agent collaboration. However, most prior works…

Code verifiers play a critical role in post-verification for LLM-based code generation, yet existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency. While reinforcement learning…

Many task-oriented dialogue systems use deep reinforcement learning (DRL) to learn policies that respond to the user appropriately and complete the tasks successfully. Training DRL agents with diverse dialogue trajectories prepare them well…

Online question-and-answer (Q\&A) systems based on the Large Language Model (LLM) have progressively diverged from recreational to professional use. This paper proposed a Multi-Agent framework with environmentally reinforcement learning…

We present SWE-Gym, the first environment for training real-world software engineering (SWE) agents. SWE-Gym contains 2,438 real-world Python task instances, each comprising a codebase with an executable runtime environment, unit tests, and…

Coding agents are increasingly used as iterative development partners, but most benchmarks still evaluate one specification followed by one final assessment. This leaves out a basic question: can an agent keep its own codebase working as…

Instruction tuning -- tuning large language models on instruction-output pairs -- is a promising technique for making models better adapted to the real world. Yet, the key factors driving the model's capability to understand and follow…

Recent advances in high-fidelity simulators have enabled closed-loop training of autonomous driving agents, potentially solving the distribution shift in training v.s. deployment and allowing training to be scaled both safely and cheaply.…