arXiv subjects

Yuyang Tian

Publications and source records attributed to Yuyang Tian.

5 recordsLinked to original sources

arXiv2026

Step-Level Preference Learning for Generative Agents in Social Simulations

Large language model (LLM)-based generative agents simulate human behavior through long-horizon decision-making processes that comprise intermediate steps such as planning, memory retrieval, reflection, and action selection. However, fine-grained human annotations of these intermediate steps remain scarce, and existing agents are not grounded in human preferences over such intermediate decisions. To address this gap, we introduce \method, an interactive simulation interface that enables us to collect step-level human preference supervision over agent decision trajectories, leading to a dataset of 57K fine-grained annotations. We conduct step-level preference learning on open-weight language models using supervised finetuning and direct preference optimization on this data, consistently improving simulation fidelity, coordination, and interaction quality, and inducing more socially effective agent behavior. Our results show that step-level human supervision is an effective training signal for improving both local decision quality and long-horizon agent behavior.

cs.AI↗

arXiv2026

ZeroWBC: Learning Natural Whole-Body Humanoid Interaction from Human Egocentric Data

Achieving versatile and natural whole-body humanoid interaction control remains challenging due to the high cost of whole-body teleoperation data. We present ZeroWBC, a teleoperation-free framework that learns humanoid whole-body interaction from human egocentric videos paired with synchronized whole-body motion and text annotations. ZeroWBC adopts a generation-then-tracking formulation to tackle the static scene whole-body interaction control problem. Given an initial egocentric image and a language instruction, a fine-tuned Vision-Language Model generates future human whole-body motion tokens, which are decoded into continuous motions and retargeted to the humanoid. The resulting reference motions, together with root and key body-part trajectories, are then executed by a general interactive motion tracking policy. To improve interaction performance, we introduce an interaction-oriented tracking reward that prioritizes global root and key body-part trajectory alignment while preserving natural whole-body motion. Experiments on the Unitree G1 humanoid robot show that ZeroWBC enables diverse scene-aware behaviors without robot teleoperation demonstrations. These results suggest a scalable paradigm for learning natural humanoid whole-body interaction from human egocentric data.

cs.RO↗

arXiv2026

Cache Your Prompt When It's Green: Carbon-Aware Caching for Large Language Model Serving

As large language models (LLMs) become widely used, their environmental impact, especially carbon emission, has attracted more attention. Prior studies focus on compute-related carbon emissions. In this paper, we find that storage is another key contributor. LLM caching, which saves and reuses KV caches for repeated context, reduces operational carbon by avoiding redundant computation. However, this benefit comes at the cost of embodied carbon from high-capacity, high-speed SSDs. As LLMs scale, the embodied carbon of storage grows significantly. To address this tradeoff, we present GreenCache, a carbon-aware cache management framework that dynamically derives resource allocation plans for LLM serving. GreenCache analyzes the correlation between carbon emission and SLO satisfaction, reconfiguring the resource over time to keep the balance between SLO and carbon emission under dynamic workloads. Evaluations from real traces demonstrate that GreenCache achieves an average carbon reduction of 15.1 % when serving Llama-3 70B in the FR grid, with reductions reaching up to 25.3 %, while staying within latency constraints for > 90 % of requests.

cs.DC↗

arXiv2025

A Visualized Framework for Event Cooperation with Generative Agents

Large Language Models (LLMs) have revolutionized the simulation of agent societies, enabling autonomous planning, memory formation, and social interactions. However, existing frameworks often overlook systematic evaluations for event organization and lack visualized integration with physically grounded environments, limiting agents' ability to navigate spaces and interact with items realistically. We develop MiniAgentPro, a visualization platform featuring an intuitive map editor for customizing environments and a simulation player with smooth animations. Based on this tool, we introduce a comprehensive test set comprising eight diverse event scenarios with basic and hard variants to assess agents' ability. Evaluations using GPT-4o demonstrate strong performance in basic settings but highlight coordination challenges in hard variants.

cs.AI↗

arXiv2023

GADY: Unsupervised Anomaly Detection on Dynamic Graphs

Anomaly detection on dynamic graphs refers to detecting entities whose behaviors obviously deviate from the norms observed within graphs and their temporal information. This field has drawn increasing attention due to its application in finance, network security, social networks, and more. However, existing methods face two challenges: dynamic structure constructing challenge - difficulties in capturing graph structure with complex time information and negative sampling challenge - unable to construct excellent negative samples for unsupervised learning. To address these challenges, we propose Unsupervised Generative Anomaly Detection on Dynamic Graphs (GADY). To tackle the first challenge, we propose a continuous dynamic graph model to capture the fine-grained information, which breaks the limit of existing discrete methods. Specifically, we employ a message-passing framework combined with positional features to get edge embeddings, which are decoded to identify anomalies. For the second challenge, we pioneer the use of Generative Adversarial Networks to generate negative interactions. Moreover, we design a loss function to alter the training goal of the generator while ensuring the diversity and quality of generated samples. Extensive experiments demonstrate that our proposed GADY significantly outperforms the previous state-of-the-art method on three real-world datasets. Supplementary experiments further validate the effectiveness of our model design and the necessity of each module.

cs.LG↗