Researcher profile

Zekun Wu

Zekun Wu contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate

Trust snapshot

Quick read

Trust 15 - UnverifiedVerification L1Unclaimed author
3works
0followers
5topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

3 published item(s)

preprint2026arXiv

NormCode: A Semi-Formal Language for Auditable AI Planning

As AI systems move into high stakes domains such as legal reasoning, medical diagnosis, and financial decision making, regulators and practitioners increasingly demand auditability. Auditability means the ability to trace exactly what each step in a multi step workflow saw and did. Current large language model based workflows are fundamentally opaque. Context pollution, defined as the accumulation of information across reasoning steps, causes models to hallucinate and lose track of constraints. At the same time, implicit data flow makes it impossible to reconstruct what any given step actually received as input. We present NormCode, a semi formal language that makes AI workflows auditable by construction. Each inference step operates in enforced data isolation and can access only explicitly passed inputs. This eliminates cross step contamination and ensures that every intermediate state can be inspected. A strict separation between semantic operations, meaning probabilistic language model reasoning, and syntactic operations, meaning deterministic data flow, allows auditors to clearly distinguish inference from mechanical restructuring. The multi format ecosystem, consisting of NCDS, NCD, NCN, and NCDN files, allows developers, domain experts, and auditors to inspect the same plan in formats suited to their individual needs. A four phase compilation pipeline transforms natural language intent into executable JSON repositories. A visual Canvas application provides real time graph visualization and breakpoint debugging. We validate the approach by achieving full accuracy on base X addition and by self hosted execution of the NormCode compiler itself. These results demonstrate that structured intermediate representations can bridge human intuition and machine rigor while maintaining full transparency.

preprint2026arXiv

Tool Calling is Linearly Readable and Steerable in Language Models

When a tool-calling agent picks the wrong tool, the failure is invisible until execution: the email gets sent, the meeting gets missed. Probing 12 instruction-tuned models across Gemma 3, Qwen 3, Qwen 2.5, and Llama 3.1 (270M to 27B), we find the identity of the chosen tool is linearly readable and steerable inside the model. Adding the mean-difference between two tools' average internal activations switches which tool the model selects at 77-100% accuracy on name-only single-turn prompts (93-100% at 4B+), and the JSON arguments that follow autoregressively match the new tool's schema, so flipping the name is enough. The same per-tool means also flag likely errors before they happen: on Gemma 3 12B and 27B, queries where the gap between the top-1 and top-2 tool is smallest produce 14-21x more wrong calls than queries with the largest gap. The causal effect concentrates along one direction, the row of the output layer that produces the target tool's first token: a unit vector along it at matched magnitude already reaches 93-100%, while what is left over leaves the choice almost untouched. Activation patching localises this to a small set of mid- and late-layer attention heads, and a within-topic probe across 14 same-domain $τ$-bench airline tools reaches top-1 61-89% across five 4B-14B models, ruling out the reading that we are just moving the model along a topic axis. Even base models encode the right tool before they can emit it: cosine readout from the internal state recovers 69-82% on BFCL while base generation reaches only 2-10%, suggesting pretraining forms the representation and instruction tuning later wires it to the output. We measure tool identity selection and JSON schema correctness in single-turn fixed-menu settings; multi-turn agentic transfer is more fragile and is discussed in Limitations.

preprint2022arXiv

Prognostic classification based on random convolutional kernel

Assessing the health status (HS) of system/component has long been a challenging task in the prognostic and health management (PHM) study. Differed from other regression based prognostic task such as predicting the remaining useful life, the HS assessment is essentially a multi class classificatIon problem. To address this issue, we introduced the random convolutional kernel-based approach, the RandOm Convolutional KErnel Transforms (ROCKET) and its latest variant MiniROCKET, in the paper. We implement ROCKET and MiniROCKET on the NASA's CMPASS dataset and assess the turbine fan engine's HS with the multi-sensor time-series data. Both methods show great accuracy when tackling the HS assessment task. More importantly, they demonstrate considerably efficiency especially compare with the deep learning-based method. We further reveal that the feature generated by random convolutional kernel can be combined with other classifiers such as support vector machine (SVM) and linear discriminant analysis (LDA). The newly constructed method maintains the high efficiency and outperforms all the other deop neutal network models in classification accuracy.