Large language model (LLM) agents are increasingly being used to automate machine learning research workflows, ranging from code development and experiment execution to research reproduction and scientific discovery, suggesting a shift from automating individual research tasks toward automating long-horizon workflows t...
Xin Liu, Hong-Ru Song, Ru-Qing Zhang et al.· 0 citations
VITAL is an importance-guided hybrid reasoning framework that selectively verbalizes informative tokens as reasoning anchors while keeping most computation latent, and achieves higher reasoning accuracy with substantially shorter reasoning trajectories on Qwen3-1.7B.
Zhe-Hao Zhou, Tong Su, Sheng-Yu Zhu et al.· 1 citation
This paper investigates the influence of LLM-generated content on term-based retrieval models, and reveals that LLM-generated texts exhibit smoother high-frequency and steeper low-frequency Zipf slopes, higher term specificity, and greater document-level diversity.
Wei Huang, Ke-Ping Bi, Yin-Qiong Cai et al.· Web Search and Data Mining· 1 citation
LLM compression reduces inference costs and memory requirements, but selecting a method and configuration remains largely empirical because comparable resource reductions can produce different capability losses. We systematically investigate capability scaling-down laws for LLM compression across pruning, quantization,...
Xue-Qi Cheng, Liang Wu, Kelly Wan et al.· 0 citations
CoCal is introduced, which trains a lightweight companion from rollout hidden states and verifier-derived correctness supervision while leaving task optimization unchanged, and shows that CoCal improves confidence estimation without sacrificing task performance, outperforming both RL-based concurrent methods and matche...
Shi-Yu Ni, Ke-Ping Bi, Jia-Feng Guo et al.· 0 citations
R$^{2}$Adapter is a lightweight plug-in Routing and Rewriting Adapter designed to allocate queries between vanilla and graph-based RAG dynamically, providing an efficient and adaptive solution for hybrid RAG systems.
Yu-Can Guo, Miao Su, Saiping Guan et al.· 0 citations
Temporal Semantic Memory is proposed, a memory framework that models semantic time for point-wise memory and supports the construction and utilization of durative memory and incorporates the query's temporal intent on the semantic timeline.
Miao Su, Yu-Can Guo, Zhong-Ni Hou et al.· Annual Meeting of the Associ...· 12 citations
An adversarially robust zero-shot image classifier that boosts the SOTA robustness by a large margin and is the first purification method in multi-modal latent space and CLIPure-Cos is the first purification method that is not based on generative models, which substantially improves defense efficiency.
Ming-Kun Zhang, Keping Bi, Wei Chen et al.· International Conference on...· 14 citations· ⚡1
HiDiffTIR is proposed, a Hierarchical Difficulty-aware policy optimization framework for multi-turn TIR that consistently improves multi-turn TIR performance and tool invocation accuracy over strong RL baselines, highlighting the necessity of difficulty-aware credit assignment for effective policy optimization in tool-...
Yu-Can Guo, Xiao-Han Wang, Miao Su et al.· 0 citations
It is shown that popularity-related signals can mitigate overconfidence and improve overall confidence estimation, and incorporate knowledge popularity reduces average confidence on incorrect answers and improves overall confidence estimation.
Shiyu Ni, Keping Bi, Jiafeng Guo et al.· 6 citations
DeepRepro dynamically transforms evolving repository states and runtime feedback into fine-grained implementation subplans, keeping planning aligned with execution throughout repository construction, and consistently outperforms strong scientific and commercial code-agent baselines.
Hong-Ru Song, Ru-Qing Zhang, Jia-Feng Guo et al.· ACM International Conference...· 1 citation
TokenWall is proposed, a runtime defense framework that acts as a semantic firewall over agent token flows, demonstrating that semantic runtime containment can achieve a practical security-utility trade-off for persistent AI agents.
Puji Wang, Yingchen Zhang, Ruqing Zhang et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.