7 papers
Break the Block: Dynamic-size Reasoning Blocks for Diffusion Large Language Models via Monotonic Entropy Descent with Reinforcement Learning
Yan Jiang, Ruihong Qiu, Zi Huang
Recent diffusion large language models (dLLMs) have demonstrated both effectiveness and efficiency in reasoning via a block-based semi-autoregressive generation paradigm. Despite t…
GFMate: Empowering Graph Foundation Models with Test-time Prompt Tuning
Yan Jiang, Ruihong Qiu, Zi Huang
Graph prompt tuning has shown great potential in graph learning by introducing trainable prompts to enhance the model performance in conventional single-domain scenarios. Recent re…
What Information Matters? Graph Out-of-Distribution Detection via Tri-Component Information Decomposition
Danny Wang, Ruihong Qiu, Zi Huang
Graph neural networks are widely used for node classification, but they remain vulnerable to out-of-distribution (OOD) shifts in node features and graph structure. Prior work estab…
Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models
Yan Jiang, Ruihong Qiu, Zi Huang
Recently, reinforcement learning (RL) has been widely applied during post-training for diffusion large language models (dLLMs) to enhance reasoning with block-wise semi-autoregress…
When to Commit? Towards Variable-Size Self-Contained Blocks for Discrete Diffusion Language Models
Danny Wang, Ruihong Qiu, Zi Huang
Discrete diffusion language models (dLLMs) enable parallel token updates with bidirectional attention, yet practical generation typically adopts blockwise semi-autoregressive decod…
TRN-R1-Zero: Text-rich Network Reasoning via LLMs with Reinforcement Learning Only
Yilun Liu, Ruihong Qiu, Zi Huang
Zero-shot reasoning on text-rich networks (TRNs) remains a challenging frontier, as models must integrate textual semantics with relational structure without task-specific supervis…