8 papers
DuoLM: Fast and Effective Personalized Recommendation with Language Models
Jiang Wu, Yutong Song, Hongbo Wang +3
Language models can make personalized recommendations from rich item content, user history, and feedback. However, evaluating many candidates with full-text inputs incurs substanti…
Cancerous Backdoor Attack: Gradual Activation via Invisible Tokens and A Stage Controller
Tahmid Imtiaz Imu, Emama Nahid, Prodipta Sen Amartya +5
Backdoor Attacks on Large Language Models (LLMs) typically follow a trigger-response pattern, where satisfying a trigger condition activates a single terminal behavior. However, th…
Evidence-Bounded Mental Health Reasoning from Heterogeneous Speech Protocols
Chengyuan Gao, Jiang Wu, Tao Lu +4
Computational mental health screening using multimodal speech and text has shown great promise. However, existing models often assume all clinical speech protocols carry equivalent…
AP-GRPO: Anchor-Gated Phonetic Alignment with Policy Optimization for Pathological Speech Reconstruction
Pengfei Zhang, Hoang H Nguyen, Yutong Song +6
Pathological speech from patients with neurodegenerative and neuromotor disorders is often acoustically distorted and linguistically fragmented, making pathological speech reconstr…
Can the Environment Speak for Itself? -GRPO: A Turn-Trajectory Group Relative Policy Optimization for Caregiver Agents
Yutong Song, Jiang Wu, Pengfei Zhang +4
Optimizing large language models (LLMs) for long-horizon caregiver agents requires balancing delayed task objectives with immediate environment dynamics, such as patient distress a…
DemMA: Dementia Multi-Turn Dialogue Agent with Expert-Guided Reasoning and Action Simulation
Yutong Song, Jiang Wu, Kazi Sharif +3
Simulating dementia patients with large language models (LLMs) is challenging due to the need to jointly model cognitive impairment, emotional dynamics, and nonverbal behaviors ove…
Do Implicit Personalization and Explicit Styles Conflict? PsPLUG: A Lightweight Plug-in for Balancing Personalization and Style in Customized LLMs
Yutong Song, Jiang Wu, Shaofan Yuan +5
Personalized large language models are often expected to follow explicit style instructions, yet we find that such instructions can undermine the user-specific characteristics that…
CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation
Yutong Song, Jiang Wu, Weijia Zhang +7
Adapting large language models to individual users remains challenging due to the tension between fine-grained personalization and scalable deployment. We present CARD, a hierarchi…