9 papers
Distill the Visual Evidence, Not Just the Answer: Cross-World On-Policy Distillation for Vision-Language Models
Yuanhao Sun, Huawei Ji, Jiaxin Ding +2
A central goal of vision-language model (VLM) distillation is to transfer both the teacher's language capabilities and its visual understanding. However, existing methods primarily…
Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn Search Agents
Yijun Zhang, Fan Xu, Jiaxin Ding +6
Reinforcement learning (RL) improves large language model (LLM) agents on long-horizon search tasks that require multiple intermediate decisions before a final outcome. However, ro…
VisPCO: Visual Token Pruning Configuration Optimization via Budget-Aware Pareto-Frontier Learning for Vision-Language Models
Huawei Ji, Yuanhao Sun, Yuan Jin +4
Visual token pruning methods effectively mitigate the quadratic computational growth caused by processing high-resolution images and video frames in vision-language models (VLMs).…
Inductive Reasoning for Temporal Knowledge Graphs with Emerging Entities
Ze Zhao, Yuhui He, Lyuwen Wu +6
Reasoning on Temporal Knowledge Graphs (TKGs) is essential for predicting future events and time-aware facts. While existing methods are effective at capturing relational dynamics,…
RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning
Bo Xue, Yuan Jin, Luoyi Fu +2
Knowledge graph reasoning (KGR) infers missing facts, with recent advances increasingly harnessing the semantic priors and reasoning abilities of Large Language Models (LLMs). Howe…
<SOG_k>: One LLM Token for Explicit Graph Structural Understanding
Jingyao Wu, Bin Lu, Zijun Di +5
Large language models show great potential in unstructured data understanding, but still face significant challenges with graphs due to their structural hallucination. Existing app…
Extreme Value Policy Optimization for Safe Reinforcement Learning
Shiqing Gao, Yihang Zhou, Shuai Shao +5
Ensuring safety is a critical challenge in applying Reinforcement Learning (RL) to real-world scenarios. Constrained Reinforcement Learning (CRL) addresses this by maximizing retur…
Controlling Underestimation Bias in Constrained Reinforcement Learning for Safe Exploration
Shiqing Gao, Jiaxin Ding, Luoyi Fu +1
Constrained Reinforcement Learning (CRL) aims to maximize cumulative rewards while satisfying constraints. However, existing CRL algorithms often encounter significant constraint v…
CHAINSFORMER: Numerical Reasoning on Knowledge Graphs from a Chain Perspective
Ze Zhao, Bin Lu, Xiaoying Gan +3
Reasoning over Knowledge Graphs (KGs) plays a pivotal role in knowledge graph completion or question answering systems, providing richer and more accurate triples and attributes. A…