9 papers
DynaKRAG: A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation
Yaqi Wu, Xiaolei Guo, Chenyu Zhou +7
Multi-hop retrieval-augmented generation (RAG) acquires evidence sequentially, with each new document potentially revealing missing facts, bridge entities, query defects, or suffic…
Dense Coordinate-List Fine-Tuning Induces a Controllable Interference Surface in Vision-Language Models
Chenyu Zhou, Qiliang Jiang, Boguang Pan
Fine-tuning vision-language models to emit dense coordinate lists improves visual grounding but also changes how models serialize, repeat, and terminate structured outputs. We stud…
LatentSkill: From In-Context Textual Skills to In-Weight Latent Skills for LLM Agents
Aofan Yu, Chenyu Zhou, Tianyi Xu +8
Agent systems increasingly use textual skills to encode reusable task procedures, but injecting these skills into the prompt at every step incurs substantial context overhead and e…
UltraVR: A Diagnostic Ultra-Resolution Image-VQA Benchmark for Evidence-Grounded Reasoning
Gexin Huang, Yanting Yang, Myeongkyun Kang +6
Vision-language models (VLMs) excel on visual question answering and multimodal reasoning benchmarks. Yet their capability on ultra-resolution images - where critical evidence is t…
OR-Space: A Full-Lifecycle Workspace Benchmark for Industrial Optimization Agents
Chenyu Zhou, Xinyun Lu, Jiangyue Zhao +3
Large language model (LLM) agents are increasingly used to assist with operations research (OR) modeling, yet existing OR-oriented benchmarks often reduce evaluation to one-shot tr…
Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents
Jingxing Wang, Chenyu Zhou, Zhihui Fu +4
Additional test-time compute can give LLM agents access to more past experience, yet expanding the context or adding rollouts does not necessarily yield greater agent capability. W…