8 papers
Information-Guided Frontier Decoding: Contextual Utility-Driven Commitment in dMLLMs
Xingyou Fang, Jingxing Zhong, Xiaosong Yuan +1
Decoding quality in diffusion multimodal language models (dMLLMs) depends heavily on the order in which masked tokens are committed. Existing confidence-based strategies prioritize…
Coalition-Aware Skill Reliability for Self-Evolving Agents
Qiyan Zhao, Xiaofeng Zhang, Bo Liu +11
Agent skills, structured artifacts distilled from interaction trajectories and dynamically reused from skill banks, have become a central mechanism for enabling large language mode…
Context-Aware Cluster Decoding: Semantic Anchor-Driven Coherence in dMLLMs
Yikai Zhao, Qiyan Zhao, Jiaquan Zhang +3
Diffusion multimodal large language models (dMLLMs) frequently produce long-form outputs marred by semantic drift and repetition, with quality generally degrading as output length…
Unlocking the Black Box of Latent Reasoning: An Interpretability-Guided Approach to Intervention
Shuochen Chang, Tong Bai, Xiaofeng Zhang +5
Latent reasoning enables Large Language Models (LLMs) to perform multi-step inference within continuous hidden states, offering efficiency gains over explicit Chain-of-Thought (CoT…
SoPE: Spherical Coordinate-Based Positional Embedding for Enhancing Spatial Perception of 3D LVLMs
Guanting Ye, Qiyan Zhao, Wenhao Yu +7
3D Large Vision-Language Models (3D LVLMs) built upon Large Language Models (LLMs) have achieved remarkable progress across various multimodal tasks. However, their inherited posit…
C^2ROPE: Causal Continuous Rotary Positional Encoding for 3D Large Multimodal-Models Reasoning
Guanting Ye, Qiyan Zhao, Wenhao Yu +4
Recent advances in 3D Large Multimodal Models (LMMs) built on Large Language Models (LLMs) have established the alignment of 3D visual features with LLM representations as the domi…
Context Tokens are Anchors: Understanding the Repetition Curse in dMLLMs from an Information Flow Perspective
Qiyan Zhao, Xiaofeng Zhang, Shuochen Chang +7
Recent diffusion-based Multimodal Large Language Models (dMLLMs) suffer from high inference latency and therefore rely on caching techniques to accelerate decoding. However, the ap…
Hallucination Begins Where Saliency Drops
Xiaofeng Zhang, Yuanchao Zhu, Chaochen Gu +8
Recent studies have examined attention dynamics in large vision-language models (LVLMs) to detect hallucinations. However, existing approaches remain limited in reliably distinguis…