13 papers
The Illusion of Visual Tool-Use: A Causal Audit of Thinking with Images
Zhiheng Wang, Bo Peng, Lai Wei +1
The "thinking-with-images" paradigm equips multimodal LLMs with active visual operations such as crop-and-zoom. However, models using these operations often achieve only marginal o…
CLBench-V: Evaluating Multimodal Context Learning from Grounding to Knowledge Acquisition
Lai Wei, Chengqi Li, Jiapeng Li +3
Real-world tasks often require models to learn from task-specific context rather than relying only on pre-trained knowledge. While recent work has highlighted this capability as co…
Bridging What the Model Thinks and How It Speaks: Expressive Speech Generation via Self-Aware Intent-Realization Alignment
Kuang Wang, Lai Wei, Ping Lin +8
Speech Language Models (SLMs) exhibit strong semantic understanding, yet often fail to translate this capacity into expressive acoustic realization, producing speech with flattened…
Attend to Evidence: Evidence-Anchored Spatial Attention Supervision for Multimodal RLVR
Ruina Hu, Chen Wang, Lai Wei +5
Reinforcement learning with verifiable rewards (RLVR) improves vision-language models (VLMs) by optimizing outcome rewards derived from final answers. However, such outcome-only re…
Targeted Exploration via Unified Entropy Control for Reinforcement Learning
Chen Wang, Lai Wei, Yanzhi Zhang +5
Recent advances in reinforcement learning (RL) have improved the reasoning capabilities of large language models (LLMs) and vision-language models (VLMs). However, the widely used…
MM-LIMA: Less Is More for Alignment in Multi-Modal Datasets
Lai Wei, Xiaozhe Li, Zihao Jiang +2
Multimodal large language models are typically trained in two stages: first pre-training on image-text pairs, and then fine-tuning using supervised vision-language instruction data…