Pingjun Hong

Papers on Lattice

Total citations

Topics

Publication activitypapers/week, last 8 weeks

Research focus

Interpretability & Mechanistic Interp (2)Natural Language Processing (1)RLHF & Preference Learning (1)Reasoning & Chain-of-Thought (1)

Frequent co-authors

Beiduo Chen (1)Ziyun Zhang (1)Benjamin Roth (1)Anna Korhonen (1)

Papers (2)

May 27, 2026

Beiduo Chen +42w ago·also Cambridge

Human Label Variation as Stable Signal: Learning Annotator-Specific Explanation Behavior via Cross-Annotator Preference Optimization

LLMs can learn to mimic the nuanced reasoning patterns of individual annotators by contrasting their explanations with those of others, opening the door to more personalized and scalable annotation strategies.

Beiduo Chen, Pingjun Hong, Ziyun Zhang +2

Interpretability & Mechanistic Interp Natural Language Processing RLHF & Preference Learning

May 25, 2026

3w ago·also ASU, UCF, UNC, Vienna

Detecting Unfaithful Chain-of-Thought via Circuit-Guided Internal-External Discrepancy

Spotting unfaithful reasoning in LLMs just got easier: a new method efficiently compares a model's internal computations against its stated rationale.

Zhen Tan, Song Wang, Pingjun Hong +2

Interpretability & Mechanistic Interp Reasoning & Chain-of-Thought

Search

Pingjun Hong

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)