MACHINE LEARNING AND INTELLIGENCE LAB

We have two papers "Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy", "Preserve or Modify? Context-Aware Evaluation for Balancing Preservation and Modification in Text-Guided Image Editing" accepted to CVPR 2025.

We have five papers "Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning", "LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding", "ReBind: Enhancing Ground-state Molecular Conformation Prediction via Force-Based Graph Rewiring", "Test-Time Ensemble via Linear Mode Connectivity: A Path to Better Adaptation", and "Token-Supervised Value Models for Enhancing Mathematical Problem-Solving Capabilities of Large Language Models" accepted to ICLR 2025.

We have one paper "LRQ: Optimizing Post-Training Quantization for Large Language Models by Learning Low-Rank Weight-Scaling Matrices" accepted to NAACL 2025.

We have two papers "Face-StyleSpeech: Enhancing Zero-shot Speech Synthesis from Face Images with Improved Face-to-Speech Mapping", "Stable-TTS: Stable Speaker-Adaptive Text-to-Speech Synthesis via Prosody Prompting" accepted to ICASSP 2025.

We have one paper "Towards Precise Prediction Uncertainty in GNNs: Refining GNNs with Topology-grouping Strategy" accepted to AAAI 2025.