Picture for Xiangyu Liu

Xiangyu Liu

Provable Partially Observable Reinforcement Learning with Privileged Information

Add code
Dec 01, 2024
Viaarxiv icon

MVBoost: Boost 3D Reconstruction with Multi-View Refinement

Add code
Nov 26, 2024
Viaarxiv icon

AdaFlow: Opportunistic Inference on Asynchronous Mobile Data with Generalized Affinity Control

Add code
Oct 31, 2024
Viaarxiv icon

fairBERTs: Erasing Sensitive Information Through Semantic and Fairness-aware Perturbations

Add code
Jul 11, 2024
Viaarxiv icon

Is poisoning a real threat to LLM alignment? Maybe more so than you think

Add code
Jun 17, 2024
Figure 1 for Is poisoning a real threat to LLM alignment? Maybe more so than you think
Figure 2 for Is poisoning a real threat to LLM alignment? Maybe more so than you think
Figure 3 for Is poisoning a real threat to LLM alignment? Maybe more so than you think
Figure 4 for Is poisoning a real threat to LLM alignment? Maybe more so than you think
Viaarxiv icon

Multi-Aspect Controllable Text Generation with Disentangled Counterfactual Augmentation

Add code
May 30, 2024
Viaarxiv icon

Do LLM Agents Have Regret? A Case Study in Online Learning and Games

Add code
Mar 25, 2024
Viaarxiv icon

Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies

Add code
Feb 20, 2024
Viaarxiv icon

Knowledge Graph Error Detection with Contrastive Confidence Adaption

Add code
Dec 19, 2023
Viaarxiv icon

Weathering Ongoing Uncertainty: Learning and Planning in a Time-Varying Partially Observable Environment

Add code
Dec 06, 2023
Viaarxiv icon