6 papers
The Decomposition Is the Fingerprint: Per-Component Identity for Agent Skills
Hongliang Liu, Yuhao Wu, Tung-Ling Li
AI agents increasingly acquire and execute skills at runtime: bundles of prompt instructions, executable code, and tool declarations fetched from marketplaces and other agents. Gov…
Leverage Is Not Reach: A Control-Window Law for Single-Neuron Steering in Language Models
Hongliang Liu
Aligned language models gate behaviors such as refusal and language routing through sparse feed forward neurons, yet no theory predicts when a single neuron intervention controls a…
AdvJudge-Zero: Binary Decision Flips in LLM-as-a-Judge via Adversarial Control Tokens
Tung-Ling Li, Yuhao Wu, Hongliang Liu
LLM-as-a-Judge systems supply the reward signal in modern RLHF and RLVR pipelines, but their binary verdict reduces to a single linear readout F_gap on one hidden state. We show th…
Breaking Safety at the Token Boundary: How BPE Tokenization Creates Exploitable Gaps in LLM Alignment
Tung-Ling Li, Hongliang Liu, Yuhao Wu
Character-level perturbations bypass safety alignment in modern LLMs despite leaving prompts human-readable. We identify and test a central structural mechanism: BPE tokenization f…
Logit-Gap Steering: A Forward-Pass Diagnostic for Alignment Robustness
Tung-Ling Li, Hongliang Liu
RLHF-style alignment trains language models to refuse unsafe requests, but how much operational margin does this refusal rest on? We introduce the refusal-affirmation logit gap: th…
Perturbation Probing: A Two-Pass-per-Prompt Diagnostic for FFN Behavioral Circuits in Aligned LLMs
Hongliang Liu, Tung-Ling Li, Yuhao Wu
Perturbation probing generates task-specific causal hypotheses for FFN neurons in large language models using two forward passes per prompt and no backpropagation, followed by a on…