Retrieval-Augmented Generation (RAG) repeatedly prefills identical text chunks across queries, incurring redundant computations. Position-Independent Caching (PIC) mitigates it by reusing precomputed Key-Value (KV) across positions, but its efficiency is constrained by the large volume of text tokens. Rendering text ch...
Yi-Lin Liu, Rui Meng, Wang-Ze Ni et al.· 0 citations
This work defines behavioral safety and introduces ActBench, a self-evolving benchmark that evaluates such behavior risk from execution trajectories rather than final responses, and proposes a dual evidence verification mechanism that verifies agent execution safety and utility through log evidence and LLM-based trajec...
H. Yao, Yimin Liu, Meihui Chen et al.· 0 citations
Text-to-image (T2I) models have substantially improved in language understanding, in-image text rendering, and visual composition, while their safety mechanisms do not always keep pace with these capabilities. This creates a cross-modal attack surface in which harmful semantics can remain inconspicuous in a serialized...
Zhi-Yi Mou, Yao Lu, Wang-Ze Ni et al.· 0 citations
Approximate nearest neighbor search (ANNS) underpins large-scale vector retrieval in search, recommendation, and retrieval-augmented generation. Graph-based indexes have demonstrated state-of-the-art search performance for ANNS. They connect each corpus vector to a small set of nearby or navigationally useful vertices...
Bao-Yuan Gu, Xiao-Yao Zhong, Jia-Bao Jin et al.· 0 citations
A new harness self-evolution method, named DREvo, is proposed, which integrates function-level evidence anchoring, state-dependent evidence recalibration, and role-conditioned search intent distillation to determine which historical evidence remains valid and where the harness should evolve next.
This work introduces a process-centric diagnostic benchmark for AI-assisted peer review that compares the decision value of Gold-process variables and Predicted-process variables, and conducts stage-level evaluation, chain-consistency evaluation, and interventional sensitivity analysis.
Siming Yuan, Xue-Yi Zhang, Wang-Ze Ni et al.· 0 citations
TAP-PER (Temporal Attentive Prefix for PERsonalization), a prefix-based framework that encodes user preferences as learnable representations, avoiding the serialization of user histories into prompts and replacing heavy per-user adapters with lightweight user-state prefix embeddings, is proposed.
Heng-Jie Cao, Fan Zhang, Jiangqi Yao et al.· arXiv.org· 1 citation
Graph-based indexes have become the dominant approach to approximate nearest neighbor search (ANNS) over high-dimensional data and play a crucial role in real-world applications such as retrieval-augmented generation, recommendation systems, and vector databases. Despite extensive progress in static graph construction...
