arXiv subjects

Zhuojun Jiang

Publications and source records attributed to Zhuojun Jiang.

6 recordsLinked to original sources

arXiv2026

Multi-Agent Closed-Loop Reasoning for Organic Structure Elucidation from Multimodal Spectra

Following the molecular discovery and synthesis revolutions, scalable automated structure elucidation from routine spectroscopic data remains an outstanding challenge. Despite decades of computational efforts, no existing system achieved reliable reasoning over unseen spectra. Here, we propose MACROS, a multi-agent system automating structure elucidation by emulating expert iterative hypothesis-testing. Trained on 100M simulated and 1.6M experimental spectra-molecule pairs, it natively supports arbitrary combinations of routine spectroscopic techniques. It achieves unprecedented zero-shot generalization to diverse real-world samples, correctly identifying synthetic compounds, natural products and metabolites above 500 Da with 1D NMR. Remarkably, MACROS spontaneously recovers textbook spectroscopic correlations from unassigned data and exhibits emergent chemical intuition such as a ring-first parsing preference, learning fundamental chemical principles rather than memorizing database patterns. MACROS augments chemists via collaboration to deliver sixfold faster, 40% more accurate elucidation. MACROS establishes a scalable foundation for fully automated structure elucidation, and catalyzes accelerated molecular discovery toward autonomous laboratories.

physics.chem-ph↗

arXiv2026

VisCanvas: A Node-Based Interface for Exploratory Visualization Authoring with LLMs

Visual data analysis involves both open-ended exploration and targeted question answering. Visualization authoring tools support this process by enabling users to create visualizations for these tasks. With the rise of large language models (LLMs), substantial effort has been devoted to developing visualization authoring tools that use natural language instructions. However, existing systems are typically based on a linear chat interface, which is not well suited to exploratory visual analysis workflows. In this paper, we introduce VisCanvas, a node-based interface for exploratory visualization authoring with LLMs. VisCanvas allows users to create, revise, branch, and merge visualizations in a non-linear way, enabling more efficient exploration of multiple analytical directions. We conducted a user study with 20 participants to evaluate the effectiveness of VisCanvas compared to a baseline chat-based interface. The results show that VisCanvas facilitates more diverse data interaction while maintaining performance levels (i.e., cognitive load and usability) that are indistinguishable from current prevailing methods. We then distill design principles for future AI-assisted visualization authoring environments. All supplemental materials required to reproduce the study are available at https://osf.io/gsxhn.

cs.HC↗

arXiv2026

HaluNet: Learning Hallucination Risk from Internal Signals in LLM Question Answering

Large language models (LLMs) achieve strong question answering (QA) performance but can produce fluent answers unsupported by available evidence. Existing hallucination detectors often rely on external verification, repeated sampling, or test-time judge calls, which can be costly for real-time QA. We propose \textbf{HaluNet}, a lightweight hallucination risk estimator that uses internal signals from one model generation. HaluNet jointly models token likelihood, predictive entropy, and hidden-state information, allowing probabilistic, distributional, and semantic evidence to inform an answer-level risk score. It is trained with LLM-as-a-Judge labels as scalable weak supervision and evaluated with independent human and multi-judge assessments. Experiments on SQuAD, TriviaQA, and Natural Questions show that HaluNet improves answer-level risk ranking across in-domain and out-of-domain settings. On a 300-example human evaluation, HaluNet achieves 0.874 AUROC and 0.869 AUPRC; its top 20\% highest-risk answers contain 96.5\% errors, yielding a 2.06$\times$ lift over the base error rate.

cs.CL↗

arXiv2023

Adversarial Demonstration Attacks on Large Language Models

With the emergence of more powerful large language models (LLMs), such as ChatGPT and GPT-4, in-context learning (ICL) has gained significant prominence in leveraging these models for specific tasks by utilizing data-label pairs as precondition prompts. While incorporating demonstrations can greatly enhance the performance of LLMs across various tasks, it may introduce a new security concern: attackers can manipulate only the demonstrations without changing the input to perform an attack. In this paper, we investigate the security concern of ICL from an adversarial perspective, focusing on the impact of demonstrations. We propose a novel attack method named advICL, which aims to manipulate only the demonstration without changing the input to mislead the models. Our results demonstrate that as the number of demonstrations increases, the robustness of in-context learning would decrease. Additionally, we also identify the intrinsic property of the demonstrations is that they can be used (prepended) with different inputs. As a result, it introduces a more practical threat model in which an attacker can attack the test input example even without knowing and manipulating it. To achieve it, we propose the transferable version of advICL, named Transferable-advICL. Our experiment shows that the adversarial demonstration generated by Transferable-advICL can successfully attack the unseen test input examples. We hope that our study reveals the critical security risks associated with ICL and underscores the need for extensive research on the robustness of ICL, particularly given its increasing significance in the advancement of LLMs.

cs.CL↗

arXiv2022

Liquid Metal Printed Ultrathin Oxides for Monolayer WS2 Top-Gate Transistors

Two-dimensional (2D) semiconductors are promising channel materials for continued downscaling of complementary metal-oxide-semiconductor (CMOS) logic circuits. However, their full potential continues to be limited by a lack of scalable high-k dielectrics that can achieve atomically smooth interfaces, small equivalent oxide thicknesses (EOT), excellent gate control, and low leakage currents. Here, we report liquid metal printed ultrathin and scalable Ga2O3 dielectric for 2D electronics and electro-optical devices. We directly visualize the atomically smooth Ga2O3/WS2 interfaces enabled by the conformal nature of liquid metal printing. We demonstrate atomic layer deposition compatibility with high-k Ga2O3/HfO2 top-gate dielectric stacks on chemical vapour deposition grown monolayer WS2, achieving EOTs of ~1 nm and subthreshold swings down to 84.9 mV/dec. Gate leakage currents are well within requirements for ultra-scaled low-power logic circuits. Our results show that liquid metal printed oxides can bridge a crucial gap in scalable dielectric integration of 2D materials for next-generation nano-electronics.

cond-mat.mtrl-sci↗

arXiv2022

Quantifying navigation complexity in transportation networks

The complexity of navigation in cities has increased with the expansion of urban areas, creating challenging transportation problems that drive many studies on the navigability of networks. However, due to the lack of individual mobility data, large-scale empirical analysis of the wayfinder's real-world navigation is rare. Here, using 225 million subway trips from three major cities in China, we quantify navigation difficulty from an information perspective. Our results reveal that 1) people conserve a small number of repeatedly used routes, and 2) the navigation information in the subnetworks formed by those routes is much smaller than the theoretical value in the global network, suggesting that the decision cost for actual trips is significantly smaller than the theoretical upper limit found in previous studies. By modeling routing behaviors in growing networks, we show that while the global network becomes difficult to navigate, navigability can be improved in subnetworks. We further present a universal linear relationship between the empirical and theoretical search information, which allows the two metrics to predict each other. Our findings demonstrate how large-scale observations can quantify real-world navigation behaviors and aid in evaluating transportation planning.

physics.soc-ph↗