Hui Cai

Hao Zhou, Siqi Cai, Hua Dai et al.

Differential privacy (DP) is crucial for safeguarding sensitive client information in federated learning (FL), yet traditional DP-FL methods rely predominantly on fixed gradient clipping thresholds. Such static clipping neglects significant client heterogeneity and varying privacy sensitivities, which may lead to an unfavorable privacy-utility trade-off. In this paper, we propose PAC-DP, a Personalized Adaptive Clipping framework for federated learning under record-level local differential privacy. PAC-DP introduces a Simulation-CurveFitting approach leveraging a server-hosted public proxy dataset to learn an effective mapping between personalized privacy budgets epsilon and gradient clipping thresholds C, which is then deployed online with a lightweight round-wise schedule. This design enables budget-conditioned threshold selection while avoiding data-dependent tuning during training. We provide theoretical analyses establishing convergence guarantees under the per-example clipping and Gaussian perturbation mechanism and a reproducible privacy accounting procedure. Extensive evaluations on multiple FL benchmarks show that PAC-DP surpasses conventional fixed-threshold approaches under matched privacy budgets, improving accuracy by up to 26% and accelerating convergence by up to 45.5% in our evaluated settings.

Yinger Zhang, Hui Cai, Xeirui Song et al.

While enabling large language models to implement function calling (known as APIs) can greatly enhance the performance of Large Language Models (LLMs), function calling is still a challenging task due to the complicated relations between different APIs, especially in a context-learning setting without fine-tuning. This paper introduces ``Reverse Chain'', a controllable, target-driven approach designed to empower LLMs with the capability to operate external APIs only via prompts. Recognizing that most LLMs have limited tool-use capabilities, Reverse Chain limits LLMs to executing simple tasks, e.g., API Selection and Argument Completion. Furthermore, to manage a controllable multi-function calling, Reverse Chain adopts a generic rule based on a backward reasoning process. This rule determines when to do API selection or Argument completion. To evaluate the multi-tool-use capability of LLMs, we have released a compositional multi-tool task dataset, available at \url{https://anonymous.4open.science/r/reverse-chain-8681}. Extensive numerical experiments validate the remarkable proficiency of Reverse Chain in managing multiple API calls.

Ibrahim Bilau, Eunhwa Yang, Hyeokhyen Kwon et al.

This study examines how visual accessibility through cabinet design influences task performance, cognitive load, physical activity level, motivation, and user experience in a virtual kitchen among older adults with and without mild cognitive impairment (MCI). Seventeen older adults (7 with MCI, 10 without) completed a repeated-measures item retrieval task under two conditions, closed cabinets and open shelving, using a counterbalanced within-subjects design. Measures included task duration, physical activity level (ENMO), cognitive load (NASA-TLX and gaze entropy), intrinsic motivation (IMI), and post-task interviews. Open shelving significantly reduced task duration (beta = -291.20, p < .001) and physical activity level (beta = -0.00615, p = .008). Gaze entropy increased (beta = 1.29, p = .001), with a significant Setting x MCI interaction (p = .009) and moderation by MoCA score (p < .001). NASA-TLX and intrinsic motivation did not differ significantly between conditions. Qualitative findings indicated reduced reliance on memory-based search and highlighted themes related to independence, aesthetics, safety, and adoption. Overall, visual accessibility improved efficiency and reduced movement demands while altering visual-search organization, with divergence between subjective and objective indicators of cognitive load. These findings support visually accessible design strategies to enhance functional performance and inform cognitively supportive built environments for aging populations.

Zhouyu Jiang, Ling Zhong, Mengshu Sun et al.

To tackle the problem of domain-specific knowledge scarcity within large language models (LLMs), knowledge graph-retrievalaugmented method has been proven to be an effective and efficient technique for knowledge infusion. However, existing approaches face two primary challenges: knowledge mismatch between public available knowledge graphs and the specific domain of the task at hand, and poor information compliance of LLMs with knowledge graphs. In this paper, we leverage a small set of labeled samples and a large-scale corpus to efficiently construct domain-specific knowledge graphs by an LLM, addressing the issue of knowledge mismatch. Additionally, we propose a three-stage KG-LLM alignment strategyto enhance the LLM's capability to utilize information from knowledge graphs. We conduct experiments with a limited-sample setting on two biomedical question-answering datasets, and the results demonstrate that our approach outperforms existing baselines.