activity
20202025
most citedSceneFormer: Indoor Scene Generation with Transformers

15 citations · 17 across the 7 of their papers we have counts for

collaborators
Showing cs.CLShow all

7 papers · 1 filter

cs.CL2025

Refusal Direction is Universal Across Safety-Aligned Languages

Xinpeng Wang, Mingyang Wang, Yihong Liu +2

Refusal mechanisms in large language models (LLMs) are essential for ensuring safety. Recent research has revealed that refusal behavior can be mediated by a single direction in ac…

cs.CL20241 cited

Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation

Xinpeng Wang, Chengzhi Hu, Paul Röttger +1

Training a language model to be both helpful and harmless requires careful calibration of refusal behaviours: Models should refuse to follow malicious instructions or give harmful…

cs.CL2024

"Seeing the Big through the Small": Can LLMs Approximate Human Judgment Distributions on NLI from a Few Explanations?

Beiduo Chen, Xinpeng Wang, Siyao Peng +3

Human label variation (HLV) is a valuable source of information that arises when multiple human annotators provide different labels for valid reasons. In Natural Language Inference…

cs.CL2024

Look at the Text: Instruction-Tuned Language Models are More Robust Multiple Choice Selectors than You Think

Xinpeng Wang, Chengzhi Hu, Bolei Ma +2

Multiple choice questions (MCQs) are commonly used to evaluate the capabilities of large language models (LLMs). One common way to evaluate the model response is to rank the candid…

cs.CL2024

"My Answer is C": First-Token Probabilities Do Not Match Text Answers in Instruction-Tuned Language Models

Xinpeng Wang, Bolei Ma, Chengzhi Hu +5

The open-ended nature of language generation makes the evaluation of autoregressive large language models (LLMs) challenging. One common evaluation approach uses multiple-choice qu…

cs.CL2023

ACTOR: Active Learning with Annotator-specific Classification Heads to Embrace Human Label Variation

Xinpeng Wang, Barbara Plank

Label aggregation such as majority voting is commonly used to resolve annotator disagreement in dataset creation. However, this may disregard minority values and opinions. Recent s…