5 papers
Tracing Query Expansion Effects through Sparse Autoencoder Features
Fangan Dong, Weiran Shi, Zhiwei Xu +5
Query expansion (QE) is a critical technique in information retrieval that enriches underspecified queries with additional textual context. However, its effect is often unreliable…
GuardianBench: A Same-Scene Instruction-Contrastive Benchmark for Latent Contextual Risk in Embodied AI
Zhesheng Zhang, Jiahao Lu, Wei Liu +8
In embodied AI, safety risk can be latent: a benign instruction and a safe scene become hazardous only when composed. Prior work has advanced embodied safety by varying visual cont…
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
Kun Yu, Jianhua Yang, Yixiang Chen +7
Language-guided human following is an important capability for embodied agents, but existing benchmarks typically assume that the target person is visible at the start of an episod…
DEEPRUBRIC: Evidence-Tree Rubric Supervision for Efficient Reinforcement Learning of Deep Research Agents
Minghang Zhu, Chuyang Wei, Junhao Xu +3
Deep research agents synthesize long-form reports by searching and reasoning over retrieved evidence. Reinforcement learning with rubric-based rewards improves these agents by opti…
MCoT-MVS: Multi-level Vision Selection by Multi-modal Chain-of-Thought Reasoning for Composed Image Retrieval
Xuri Ge, Chunhao Wang, Xindi Wang +3
Composed Image Retrieval (CIR) aims to retrieve target images based on a reference image and modified texts. However, existing methods often struggle to extract the correct semanti…