arXiv Computation and Language

XQDT: eXplainable and Quantitative Data-Text Alignment Metric with Feedback Signals

arXiv Computation and Language
Sep 2

Post-hoc Alignment of LLM-judges to Human Judgment Distribution

The paper introduces NAPHA, a lightweight post‑hoc alignment method that improves large language model (LLM) predictions of human judgment distributions (HJD) by matching LLM output distributions to HJD through entropy‑based class assignment and specialized alignment models. Experiments on five datasets show that while LLMs perform near human‑level on hard‑label tasks, they struggle with soft‑label predictions, and NAPHA consistently enhances soft‑label accuracy, especially on high‑entropy instances. The study also demonstrates that better entropy class prediction can further boost NAPHA’s effectiveness.

By Sebastian Steindl, Nikos Voskarides, Alberto Gasparin, Diego Marcheggiani
arXiv AI
Oct 9

Predicting Alignment Generalization with Value Representations

The paper introduces the task of predicting alignment generalization, which measures how fine‑tuning a language model to follow a specific value affects its behavior across a broad set of other values. Through a large‑scale study of 66 values, the authors show that representations derived from model activations when applying values in context outperform textual description‑based methods, achieving a correlation of 0.45 versus 0.05. They further demonstrate that these representations can predict downstream robustness and help construct a taxonomy of LLM values based on empirical generalization dynamics.

By Andy Liu, Mehar Bhatia, Karolina Stanczak, Mona Diab, Vered Shwartz, Daniel Fried
arXiv AI
Sep 3

Automated Researchers Can Mitigate Well-characterized Alignment Failures

The paper investigates whether automated alignment researchers (AARs) can post‑train language models to reduce well‑characterized alignment failures such as deception, sycophancy, and jailbreaks while preserving general capability. Across ten failures, the strongest AAR methods significantly lower targeted failures and generalize to held‑out benchmarks, larger models, and multi‑turn audits. In contrast, a human baseline of 28 experienced researchers, given eight hours to devise one‑shot methods, underperformed the best AAR approaches, and providing human ideas to AARs did not improve results.

By Chen Yueh-Han, Jiaxin Wen, Jan Hendrik Kirchner
arXiv Computation and Language
Oct 16

VALCON: Benchmarking and Enhancing Value Consistency in Large Language Models via Dual-Role Variation

VALCON introduces a role-driven framework to benchmark and improve value consistency in large language models (LLMs) by evaluating how well an LLM’s generative outputs align with its evaluative judgments when the same model alternates between generator and evaluator roles. The framework creates three benchmark suites—VAB-Consistency, VAB-Boundary, and VAB-Challenge—using dual-role variation to expose hidden value discrepancies. Experiments across 28 LLMs reveal widespread cross-role inconsistency, and mechanistic probing identifies a role-dependent representational decoupling; a representation-directed intervention is then proposed to mitigate these failures, establishing value consistency as a distinct alignment dimension.

By Junru Wu, Jingting Zheng, Yuqi Ren, Deyi Xiong
arXiv AI
Oct 31

Reliability of LLM Judges for Evaluating Entity Alignment

The paper investigates how reliable large language models (LLMs) are when used as judges to evaluate entity alignment (EA) systems, which match equivalent entities across knowledge graphs. Through systematic benchmarking across three frontier LLMs, three datasets, and four EA systems, the study identifies a failure mode called anchor bias, where judges invert discrimination when the system’s decision label is visible. The authors show that a label‑free evaluation protocol restores high discrimination accuracy, while visible labels cause significant score shifts, especially in stronger models—a phenomenon they term the frontier model paradox. They also provide a new biomedical EA benchmark (MeSH–SNOMED CT) and an open‑source framework for auditing LLM judge reliability.

By Vaibhava Lakshmi Ravideshik, Mayank Kejriwal
arXiv AI
Sep 30

Alignment Forecasting: Predicting Misalignment From Training Data

The paper introduces Alignment Forecasting, a method for predicting whether fine‑tuning a language model on a given dataset will increase specific alignment failures such as deception or sycophancy. It presents ALIGNMENTFORECASTBENCH, a benchmark of over 5,000 forecasting questions across many models, datasets, and failure modes, and shows that a simple forecasting scaffold using an LLM’s assessment of dataset bias can outperform baseline forecasters. The authors demonstrate that filtering out high‑risk training examples identified by the forecaster can improve alignment in multiple‑choice evaluations, though benefits in open‑ended conversations remain uncertain.

By Chen Yueh-Han, Bruce W. Lee, Ilia Sucholutsky, Tomek Korbak