24 papers
Uncovering Semantic Hierarchies in Text-Attributed Graphs via Variational EM-based LLM-GNN Synergy
Yunhui Liu, Xudong Jin, Qizhuo Xie +5
While hierarchical structures are intrinsic to organizing human knowledge, current representation learning on text-attributed graphs predominantly operates on flat semantic spaces,…
The Window Is in the Reasoning: When LLM Agents Decide to Follow Prompt Injections
Haoxuan Jia, Yang Liu, Bin Chong +13
LLM agents that read untrusted content can be hijacked by prompt injections. Runtime monitors that try to catch a hijack before the agent acts, and benchmarks that score each task…
CARAT: Do Materials LLMs Reason or Recite?
Jiajun Wu, Jian Yang, Zixiang Ni +2
When a materials LLM answers a question about crystal structure, does it reason from the structure or copy an answer already printed in its input? Accuracy cannot tell: a structura…
SWE-PolyVision: Benchmarking Cross-Image Abductive Reasoning for Repository-Level Software Engineering
Jiajun Wu, Leixin Sun, Zihan Tan +12
Current multimodal software-engineering benchmarks expose images as additional context, but do not test whether an agent can integrate evidence distributed across images into a ver…
SWE-Prometheus: Measuring Engineering Governance Improvements in Real-World Repositories
Jiajun Wu, Leixin Sun, Zihan Tan +12
Large language model based coding agents have made substantial progress on repository-level software engineering tasks. Existing repository benchmarks, however, usually start from…
AquaMend: Minimal Re-probing and Conditional Rollback for Latent-Belief Failures in Embodied Agents
Yufan Liu, Shang Luo, Yang Liu +9
Physical changes or sensing errors can invalidate embodied agents' task-relevant beliefs. AquaMend compares re-probing, rollback, and supported continuation on a probe-belief-actio…
Granularity-Adaptive Credit Assignment for Long-Horizon LLM Agent Reinforcement Learning
Taoran Liang, Yang Liu, Shang Luo +10
Long-horizon language-model agents trained with reinforcement learning oftenreceive sparse outcome rewards that do not reveal which decisions along a tra-jectory deserve credit. Ep…
Harness-agnostic detection and immunization of reward hacking in self-evolving language models
Rongxin Yang, Yang Liu, Shang Luo +10
Self-evolving language models improve by proposing candidate updates and keeping whatever raises a visible score. When that score is an imperfect proxy for the capability one actua…
Safety Does Not Compose: Non-Decaying Loop State for Autonomous LLM Agents
Chenhao Wu, Haoxuan Jia, Yang Liu +11
Large language model agents are increasingly deployed as autonomous loops. Starting from one human goal, such a system repeatedly discovers work, plans, executes tool calls, verifi…
Read Between the Stickers: Sentiment-Prior Reasoning with Learnable Verbalized Rules for Multimodal Chat Analysis
Zixiang Ni, Yifei Xu, Haowen Yang +8
Multimodal chat analysis of social media stickers (MCAS) benefits from jointly modeling text and sticker semantics, yet it is inherently challenged by the interference between sent…
Hindsight Memory-PRM: Supervising Memory Management with Auditable Hindsight Credit
Haoxuan Jia, Yang Liu, Yingguang Yang +14
Memory operations of long-horizon LLM agents are hard to supervise: an operation's value is unobservable when it is taken. But they are special -- they leave machine-readable evide…
Certified Multi-Turn Robustness for LLM Safety via Compositional Bounds and Safety Persistence
Yang Liu, Bin Chong, Wenkai Yang +9
Large language models (LLMs) are vulnerable to multi-turn jailbreak attacks that progressively manipulate conversation context. Existing certified robustness methods are limited to…
DiffImaginE: Imagine to Verify Entity Types with Diffusion
Feng Zhang, Feiyu Han, Rongxin Yang +11
Multimodal named entity recognition (MNER) determines whether each candidate span and entity-type hypothesis is supported by joint textual and visual evidence. Existing imagine-and…
Physiological World Models for Human State Transitions
Chongyang Zhang, Rendong Wang, Hao Zheng +4
Continuous multimodal sensing now allows human physiology to be observed throughout daily life rather than only during occasional clinical visits. However, most health artificial i…
Thought-Aware KV Cache Compaction for Reasoning via Adaptive Attention Matching
Yang Liu, Bin Chong, Chongyang Zhang +3
Reasoning language models generate lengthy chain-of-thought (CoT) sequences whose key-value (KV) cache grows linearly and becomes a memory bottleneck during decoding. Existing comp…
SAHG: Sector-Anisotropic Hyperbolic Graph Model for Social Bot Detection
Hanning Lu, Yingguang Yang, Jinwei Su +8
LLM-driven social bots can generate fluent, human-like text, reducing the discriminative advantage of content-based detection alone. However, coordinated campaigns still leave rela…
FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents
Haoxuan Jia, Yang Liu, Bin Chong +10
Finance LLM agents must simultaneously block prompt-induced unauthorized actions and approve legitimate multi-step business workflows. However, boundary filters often miss irrevers…
ExTax: Explainable Disinformation Detection via Persuasion, Emotion, and Narrative Role Taxonomies
Shang Luo, Yingguang Yang, Zhenchen Sun +8
The democratization of LLMs has accelerated the generation and circulation of highly fluent disinformation, making traditional syntax-semantic verification increasingly insufficien…
SafeHarness: Lifecycle-Integrated Security Architecture for LLM-based Agent Deployment
Xixun Lin, Yang Liu, Yancheng Chen +8
The performance of large language model (LLM) agents depends critically on the execution harness, the system layer that orchestrates tool use, context management, and state persist…
Reducing Detail Hallucinations in Long-Context Regulatory Understanding via Targeted Preference Optimization
Yang Liu, Bin Chong, Yuhan Lin +7
Large language models (LLMs) frequently produce \emph{detail hallucinations} when processing long regulatory documents, including subtle errors in threshold values, units, scopes,…
FedRio: Personalized Federated Social Bot Detection via Cooperative Reinforced Contrastive Adversarial Distillation
Yingguang Yang, Hao Liu, Xin Zhang +8
Social bot detection is critical to the stability and security of online social platforms. However, current state-of-the-art bot detection models are largely developed in isolation…
Mitigating Homophily Disparity in Graph Anomaly Detection: A Scalable and Adaptive Approach
Yunhui Liu, Qizhuo Xie, Yinfeng Chen +4
Graph anomaly detection (GAD) aims to identify nodes that deviate from normal patterns in structure or features. While recent GNN-based approaches have advanced this task, they str…
RoBCtrl: Attacking GNN-Based Social Bot Detectors via Reinforced Manipulation of Bots Control Interaction
Yingguang Yang, Xianghua Zeng, Qi Wu +5
Social networks have become a crucial source of real-time information for individuals. The influence of social bots within these platforms has garnered considerable attention from…
User Trajectory Prediction Unifying Global and Local Temporal Information
Wei Hao, Bin Chong, Ronghua Ji +1
Trajectory prediction is essential for formulating proactive strategies that anticipate user mobility and support advance preparation. Therefore, how to reduce the forecasting erro…