English
Related papers

Related papers: Debate Helps Weak Judges Reward Stronger Models

200 papers

To make AI systems broadly useful for challenging real-world tasks, we need them to learn complex human goals and preferences. One approach to specifying complex goals asks humans to judge during training which agent behaviors are safe and…

Machine Learning · Statistics 2018-10-23 Geoffrey Irving , Paul Christiano , Dario Amodei

We propose a simple model to explore an educational phenomenon where the correct answer emerges from group discussion. We construct our model based on several plausible assumptions: (i) We tend to follow peers' opinions. However, if a…

Physics and Society · Physics 2025-05-27 Jibeom Seo , Beom Jun Kim

Developing tools to automatically detect check-worthy claims in political debates and speeches can greatly help moderators of debates, journalists, and fact-checkers. While previous work on this problem has focused exclusively on the text…

Computation and Language · Computer Science 2024-01-19 Petar Ivanov , Ivan Koychev , Momchil Hardalov , Preslav Nakov

The deliberative potential of online platforms has been widely examined. However, little is known about how various interface-based reflection nudges impact the quality of deliberation. This paper presents two user studies with 12 and 120…

Human-Computer Interaction · Computer Science 2025-02-07 Shun Yi Yeo , Gionnieve Lim , Jie Gao , Weiyu Zhang , Simon Tangi Perrault

In many domains of life, business and management, numerous problems are addressed by small groups of individuals engaged in face-to-face discussions. While research in social psychology has a long history of studying the determinants of…

Physics and Society · Physics 2018-02-07 Mehdi Moussaid , Alejandro Noriega Campero , Abdullah Almaatouq

An audience's prior beliefs and morals are strong indicators of how likely they will be affected by a given argument. Utilizing such knowledge can help focus on shared values to bring disagreeing parties towards agreement. In argumentation…

Computation and Language · Computer Science 2022-03-29 Milad Alshomary , Roxanne El Baff , Timon Gurcke , Henning Wachsmuth

Weighted abduction computes hypotheses that explain input observations. A reasoner of weighted abduction first generates possible hypotheses and then selects the hypothesis that is the most plausible. Since a reasoner employs parameters,…

Logic in Computer Science · Computer Science 2025-02-17 Shota Motoura , Ayako Hoshino , Itaru Hosomi , Kunihiko Sadamasa

Peer reviewing is a central process in modern research and essential for ensuring high quality and reliability of published work. At the same time, it is a time-consuming process and increasing interest in emerging fields often results in a…

Computers and Society · Computer Science 2020-12-15 Michael Fromm , Evgeniy Faerman , Max Berrendorf , Siddharth Bhargava , Ruoxia Qi , Yao Zhang , Lukas Dennert , Sophia Selle , Yang Mao , Thomas Seidl

Recently, there has been a trend of evaluating the Large Language Model (LLM) quality in the flavor of LLM-as-a-Judge, namely leveraging another LLM to evaluate the current output quality. However, existing judges are proven to be biased,…

Computation and Language · Computer Science 2024-09-26 Hongli Zhou , Hui Huang , Yunfei Long , Bing Xu , Conghui Zhu , Hailong Cao , Muyun Yang , Tiejun Zhao

State-of-the-art single-agent claim verification methods struggle with complex claims that require nuanced analysis of multifaceted evidence. Inspired by real-world professional fact-checkers, we propose \textbf{DebateCV}, the first…

Computation and Language · Computer Science 2026-04-06 Haorui He , Yupeng Li , Dacheng Wen , Yang Chen , Reynold Cheng , Donglong Chen , Francis C. M. Lau

This work extends a model of simulating influence in a network of stochastic edge dynamics to account for polarization. The model built upon is termed Dynamic Communicators and seeks to understand the process which produces low volume, high…

Social and Information Networks · Computer Science 2018-05-29 Cameron E. Taylor , Ivan Garibay , Alexander V. Mantzaris

In a world where ideas flow freely between people across multiple platforms, we often find ourselves relying on others' information without an objective standard to judge whether those opinions are accurate. The present study tests an…

Social and Information Networks · Computer Science 2019-03-27 Niccolo Pescetelli , Nick Yeung

Reward schemes may affect not only agents' effort, but also their incentives to gather information to reduce the riskiness of the productive activity. In a laboratory experiment using a novel task, we find that the relationship between…

General Economics · Economics 2024-09-11 Philip Brookins , Jennifer Brown , Dmitry Ryvkin

Multi-judge evaluation is increasingly used to assess LLMs and reward models, and the prevailing heuristic is to curate: keep the most accurate judges and discard weaker ones. We show that this heuristic can reverse when the target is not…

Methodology · Statistics 2026-05-12 Yanran Li

The $\textit{LLM-as-a-judge}$ paradigm has become the operational backbone of automated AI evaluation pipelines, yet rests on an unverified assumption: that judges evaluate text strictly on its semantic content, impervious to surrounding…

Artificial Intelligence · Computer Science 2026-04-17 Manan Gupta , Inderjeet Nair , Lu Wang , Dhruv Kumar

Large language models (LLMs) has been widely adopted as a scalable surrogate for human evaluation, yet such judges remain imperfect and susceptible to surface-level biases. One possible reason is that these judges lack sufficient…

Computation and Language · Computer Science 2026-04-09 Minzhu Tu , Shiyu Ni , Keping Bi

As an epistemic activity, rational debate and discussion requires cooperation, yet involves a tension between collective and individual interests. While all participants benefit from collective outcomes like reaching consensus on true…

Social and Information Networks · Computer Science 2025-04-10 Toby Handfield , Julián Garcia , Christian Hilbe , Shang Long Yeo

In-depth analysis of competitive debates is essential for participants to develop argumentative skills and refine strategies, and further improve their debating performance. However, manual analysis of unstructured and unlabeled textual…

Human-Computer Interaction · Computer Science 2026-01-07 Qianhe Chen , Yong Wang , Yixin Yu , Xiyuan Zhu , Xuerou Yu , Ran Wang

Tabular anomaly detection is often handled by single detectors or static ensembles, even though strong performance on tabular data typically comes from heterogeneous model families (e.g., tree ensembles, deep tabular networks, and tabular…

Machine Learning · Computer Science 2026-02-17 Pinqiao Wang , Sheng Li

Self-supervision provides effective representations for downstream tasks without requiring labels. However, existing approaches lag behind fully supervised training and are often not thought beneficial beyond obviating or reducing the need…

Machine Learning · Computer Science 2019-10-30 Dan Hendrycks , Mantas Mazeika , Saurav Kadavath , Dawn Song