English
Related papers

Related papers: The Alignment Game: A Theory of Long-Horizon Align…

200 papers

Current human-AI alignment and evaluation methods for large language models (LLMs) often rely on preference signals collected immediately after an interaction. This practice implicitly treats preference as static, even though many…

Human-Computer Interaction · Computer Science 2026-05-06 Simret Araya Gebreegziabher , Allison E Sproul , Yinuo Yang , Chaoran Chen , Diego Gómez-Zará , Toby Jia-Jun Li

It has recently been argued that AI models' representations are becoming aligned as their scale and performance increase. Empirical analyses have been designed to support this idea and conjecture the possible alignment of different…

Machine Learning · Computer Science 2025-02-21 Francesco Insulla , Shuo Huang , Lorenzo Rosasco

Standard methods for aligning large language models with human preferences learn from pairwise comparisons among sampled candidate responses and regularize toward a reference policy. Despite their effectiveness, the effects of sampling and…

Machine Learning · Computer Science 2026-02-13 Yurong Chen , Yu He , Michael I. Jordan , Fan Yao

Existing multi-objective reinforcement learning (MORL) algorithms do not account for objectives that arise from players with differing beliefs. Concretely, consider two players with different beliefs and utility functions who may cooperate…

Artificial Intelligence · Computer Science 2017-05-16 Andrew Critch

Emerging research in Pluralistic Artificial Intelligence (AI) alignment seeks to address how intelligent systems can be designed and deployed in accordance with diverse human needs and values. We contribute to this pursuit with a dynamic…

Machine Learning · Computer Science 2024-11-01 Hadassah Harland , Richard Dazeley , Peter Vamplew , Hashini Senaratne , Bahareh Nakisa , Francisco Cruz

Machine Learning (ML) increasingly informs the allocation of opportunities to individuals and communities in areas such as lending, education, employment, and beyond. Such decisions often impact their subjects' future characteristics and…

Machine Learning · Computer Science 2021-06-03 David Lindner , Hoda Heidari , Andreas Krause

A pervasive challenge in Reinforcement Learning (RL) is the "curse of dimensionality" which is the exponential growth in the state-action space when optimizing a high-dimensional target task. The framework of curriculum learning trains the…

Machine Learning · Computer Science 2025-03-24 Mingxuan Li , Junzhe Zhang , Elias Bareinboim

Alignment faking is a form of strategic deception in AI in which models selectively comply with training objectives when they infer that they are in training, while preserving different behavior outside training. The phenomenon was first…

Large language models (LLMs) are currently aligned using techniques such as reinforcement learning from human feedback (RLHF). However, these methods use scalar rewards that can only reflect user preferences on average. Pluralistic…

Computation and Language · Computer Science 2025-08-13 Jadie Adams , Brian Hu , Emily Veenhuis , David Joy , Bharadwaj Ravichandran , Aaron Bray , Anthony Hoogs , Arslan Basharat

Reward learning plays a pivotal role in Reinforcement Learning from Human Feedback (RLHF), ensuring the alignment of language models. The Bradley-Terry (BT) model stands as the prevalent choice for capturing human preferences from datasets…

Machine Learning · Computer Science 2024-10-10 Jinsong Liu , Dongdong Ge , Ruihao Zhu

We examine the tuning of cooperative behavior in repeated multi-agent games using an analytically tractable, continuous-time, nonlinear model of opinion dynamics. Each modeled agent updates its real-valued opinion about each available…

Physics and Society · Physics 2021-11-24 Shinkyu Park , Anastasia Bizyaeva , Mari Kawakatsu , Alessio Franci , Naomi Ehrich Leonard

Pluralistic alignment is typically operationalised as preference aggregation: producing responses that span (Overton), steer toward (Steerable), or proportionally represent (Distributional) diverse human values. We argue that aggregation…

Artificial Intelligence · Computer Science 2026-05-15 Varad Vishwarupe , Nigel Shadbolt , Marina Jirotka

The tendency of repeating past choices more often than expected from the history of outcomes has been repeatedly empirically observed in reinforcement learning experiments. It can be explained by at least two computational processes:…

Neural and Evolutionary Computing · Computer Science 2024-10-28 Isabelle Hoxha , Leo Sperber , Stefano Palminteri

Aligning large language models (LLMs) with human intent is critical for enhancing their performance across a variety of tasks. Standard alignment techniques, such as Direct Preference Optimization (DPO), often rely on the binary…

Computation and Language · Computer Science 2025-04-01 Yuxiang Guo , Lu Yin , Bo Jiang , Jiaqi Zhang

Currently, most reinforcement learning tasks focus on domains like mathematics and programming, where verification is relatively straightforward. However, in subjective tasks such as role-playing, alignment techniques struggle to make…

After pre-training, large language models are aligned with human preferences based on pairwise comparisons. State-of-the-art alignment methods (such as PPO-based RLHF and DPO) are built on the assumption of aligning with a single preference…

Machine Learning · Computer Science 2025-05-30 Paul Gölz , Nika Haghtalab , Kunhe Yang

Reinforcement learning from human feedback (RLHF) implicitly aggregates heterogeneous human preferences into a single utility function, even though the underlying utilities of the participants are in practice diverse. Hence, RLHF can be…

Computer Science and Game Theory · Computer Science 2026-01-28 Zhiyu An , Duaa Nakshbandi , Wan Du

Reward models are central to Large Language Model (LLM) alignment within the framework of RLHF. The standard objective used in reward modeling is the Bradley-Terry (BT) loss, which learns from pairwise data consisting of chosen and rejected…

Machine Learning · Computer Science 2026-02-03 Tong Xie , Andrew Bai , Yuanhao Ban , Yunqi Hong , Haoyu Li , Cho-jui Hsieh

In this work, we consider policy-based methods for solving the reinforcement learning problem, and establish the sample complexity guarantees. A policy-based algorithm typically consists of an actor and a critic. We consider using various…

Machine Learning · Computer Science 2023-01-16 Zaiwei Chen , Siva Theja Maguluri

In the context of reinforcement learning from human feedback (RLHF), the reward function is generally derived from maximum likelihood estimation of a random utility model based on pairwise comparisons made by humans. The problem of learning…

Computer Science and Game Theory · Computer Science 2024-11-08 Luise Ge , Daniel Halpern , Evi Micha , Ariel D. Procaccia , Itai Shapira , Yevgeniy Vorobeychik , Junlin Wu