collaborators
Showing cs.CLShow all

9 papers · 1 filter

cs.CL2026

SemTrace: Source-Grounded Semantic Signatures for Tracing LLM Exposure to Protected Documents

Junyan Zhang, Yudong Zeng, Yongwei Huang +3

Large language models are increasingly used to read documents and produce downstream text, creating a provenance problem when the document owner cannot control or inspect the model…

cs.CL2026

Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation

Sihang Jia, Shuliang Liu, Songbo Yang +3

Multimodal Large Language Models frequently suffer from inference hallucinations, partially stemming from language priors dominating visual evidence. Existing training-free mitigat…

cs.CL2026

Correct Prediction, Wrong Steps? Consensus Reasoning Knowledge Graph for Robust Chain-of-Thought Synthesis

Zipeng Ling, Shuliang Liu, Seonil Son +4

Large language models (LLMs) have become increasingly used for various tasks, often coupled with Chain-of-Thought (CoT) prompting to boost accuracy. Recent work has shown that high…

cs.CL2025

Quantifying LLM Biases Across Instruction Boundary in Mixed Question Forms

Zipeng Ling, Shuliang Liu, Yuehao Tang +8

Large Language Models (LLMs) annotated datasets are widely used nowadays, however, large-scale annotations often show biases in low-quality datasets. For example, Multiple-Choice Q…

cs.CL2025

LLM Abstention Can Be a Prompt Artifact, in Addition to Genuine Uncertainty

Zipeng Ling, Shuliang Liu, Yuehao Tang +8

Large Language Models (LLMs) are increasingly trained to abstain from answering questions they are unsure about. However, this ability is often misused: in real-world applications,…

cs.CL2025

Unveiling Instruction-Specific Neurons & Experts: An Analytical Framework for LLM's Instruction-Following Capabilities

Junyan Zhang, Yubo Gao, Yibo Yan +8

The finetuning of Large Language Models (LLMs) has significantly advanced their instruction-following capabilities, yet the underlying computational mechanisms driving these improv…