LLM Alignment

LLM: Large Language Model

LLM alignment studies how to shape language models’ behavior so they are helpful, truthful, safe, and responsive to the people and settings they serve. Common approaches include supervised fine-tuning, preference learning such as Direct Preference Optimization, reinforcement learning from human or AI feedback, and inference-time steering or retrieval. Researchers evaluate these methods on instruction-following, preference, jailbreak, factuality, multilingual, and multi-turn tasks, while also developing ways to represent disagreement among users and communities. A central challenge is that gains in safety or preference alignment may reduce usefulness, diversity, calibration, or capability, and desired behavior may not generalize to new prompts, groups, or deployment settings.

Latest papers

613 papers58 papers in the last four weeks, up 12% on the four weeks before. 0.3% of all new papers.

Pressure Reveals Character: Behavioural Alignment Evaluation at Depth
Feb 24, 2026Nora Petrova, John BurdenLLM Alignment LLM Safety Benchmarks
Prolific
PromptCD: Test-Time Behavior Enhancement via Polarity-Prompt Contrastive Decoding
Feb 24, 2026Baolong Bi, Yuyao Ge, Shenghua Liu +9Inference-Time Language Model Alignment LLM Alignment
Key Laboratory of Network Data Science and Technology, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China · University of Chinese Academy of Sciences · University of Chinese Academy of Sciences, Beijing, China +3
CAMEL: Confidence-Gated Reflection for Reward Modeling
Feb 24, 2026Zirui Zhu, Hailun Xu, Yang Luo +4Reward Modeling LLM Alignment
National University of Singapore · TikTok
Oracle-Robust Online Alignment for Large Language Models
Feb 24, 2026Zimeng Li, Mudit Gaur, Vaneet AggarwalLLM Alignment Preference-Based RL
Purdue University
AI Alignment Breaks at the Edge
Feb 23, 2026Han Bao, Yue Huang, Xiaoda Wang +5LLM Alignment AI Safety Evaluation
University of Notre Dame · Emory University
Soft Sequence Policy Optimization
Feb 22, 2026Svetlana Glazyrina, Maksim Kryzhanovskiy, Roman IschenkoLLM Alignment RL Post-Training
Lomonosov Moscow State University, Moscow, Russia · Institute for Artificial Intelligence, Lomonosov Moscow State University, Moscow, Russia
Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models
Feb 22, 2026Seong Hah Cho, Junyi Li, Anna LeshinskayaLLM Alignment Moral Reasoning in Language Models
Independent · Department of Cognitive Sciences, UC Irvine · Department of Cognitive Sciences, UC Irvine & AI Objectives Institute
ODESteer: A Unified ODE-Based Steering Framework for LLM Alignment
Feb 19, 2026Hongjue Zhao, Haosen Sun, Jiangtao Kong +8Language Model Steering Activation Steering
University of Illinois Urbana-Champaign · Northwestern University · William & Mary +2
Fail-Closed Alignment for Large Language Models
Feb 19, 2026Zachary Coalson, Beth Sohler, Aiden Gabriel +1LLM Alignment Refusal Behavior in Language Models
Oregon State University, Corvallis, OR USA.
References Improve LLM Alignment in Non-Verifiable Domains
Feb 18, 2026Kejian Shi, Yixin Liu, Peifeng Wang +3LLM Alignment LLM-as-a-Judge
Yale University · Meta · Scale AI +2
Aligning Language Models from User Interactions
Feb 18, 2026Thomas Kleine Buening, Jonas Hübotter, Barna Pásztor +3LLM Alignment Personalized Language Model Alignment
ETH Zurich · MIT · University of Zurich
Intra-Fairness Dynamics: The Bias Spillover Effect in Targeted LLM Alignment
Feb 18, 2026Eva Paraschou, Line Harder Clemmensen, Sneha DasLLM Alignment Language Model Bias Evaluation
Department of Applied Mathematics and Computer Science, Technical University of Denmark, 2800 Lyngby, Denmark · Department of Mathematical Sciences, University of Copenhagen, 2100 Copenhagen, Denmark
Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment
Feb 17, 2026Jialu Wang, Heinrich Peters, Asad A. Butt +6LLM Alignment Group Relative Policy Optimization
1Apple Inc., California, United States.
Operationalising the Superficial Alignment Hypothesis via Task Complexity
Feb 17, 2026Tomás Vergara-Browne, Darshan Patil, Ivan Titov +3LLM Alignment Language Model Post-Training
Mila Quebec AI Institute · McGill University · Université de Montréal +4
The Geometry of Alignment Collapse: When Fine-Tuning Breaks Safety
Feb 17, 2026Max Springer, Chung Peng Lee, Blossom Metevier +5LLM Safety Alignment LLM Alignment
Princeton University
Alignment as Iatrogenesis: Pastoral Power, Collective Pathology, and the Structural Limits of Monolingual Safety Evaluation
Feb 17, 2026Hiroki FukuiLLM Alignment Language Model Safety Evaluation
Department of Neuropsychiatry, Graduate School of Medicine, Kyoto University
Discovering Implicit Large Language Model Alignment Objectives
Feb 17, 2026Edward Chen, Sanmi Koyejo, Carlos GuestrinLLM Alignment LLM Interpretability
Stanford University.
Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment
Feb 16, 2026Laurène Vaugrante, Anietta Weckauff, Thilo HagendorffLLM Alignment Language Model Self-Assessment
Interchange Forum for Reflecting on Intelligent Systems, University of Stuttgart, Stuttgart, Germany
Socially-Weighted Alignment: A Game-Theoretic Framework for Multi-Agent LLM Systems
Feb 16, 2026Furkan Mumcu, Yasin YilmazCongestion Games Multi-Agent LLM Systems
University of South Florida
Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges
Feb 14, 2026Ruomeng Ding, Yifei Pang, He Sun +3LLM Alignment LLM-as-a-Judge
University of North Carolina at Chapel Hill · 2Carnegie Mellon University · 3Yale University +1
Mitigating the Safety-utility Trade-off in LLM Alignment via Adaptive Safe Context Learning
Feb 14, 2026Yanbo Wang, Minzheng Wang, Jian Liang +3LLM Alignment RL for Language Model Reasoning
School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China. · NLPR & MAIS, Institute of Automation, Chinese Academy of Sciences, Beijing, China. · Ritzz-AI.
Learning Ordinal Probabilistic Reward from Preferences
Feb 13, 2026Longze Chen, Qi Deng, Lu Wang +7Reward Modeling LLM Alignment
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Ritzz-AI +1
Evaluating Alignment of Behavioral Dispositions in LLMs
Feb 11, 2026Amir Taubenfeld, Zorik Gekhman, Lior Nezry +8Human Preference Evaluation LLM Alignment
Google Research · Hebrew University · University of Cambridge
Routing-Aware Safety Alignment for Mixture-of-Experts Models
Feb 4, 2026Jiacheng Liang, Yuhui Wang, Tanqiu Jiang +1Mixture-of-Experts Language Models LLM Alignment
Department of Computer Science Stony Brook University
Learning Where It Matters: Geometric Anchoring for Robust Preference Alignment
Feb 4, 2026Youngjae Cho, Jongsuk Kim, Ji-Hoon KimLLM Alignment Preference Alignment
PYLER · KAIST
Towards Optimality in Active Preference Learning
Feb 2, 2026Yao Zhao, Kwang-Sung JunActive Preference Learning Active Learning
Department of Computer Science, University of Arizona, Tucson, USA.
Continuous-Utility Direct Preference Optimization
Jan 31, 2026Muhammad Ahmed Mohsin, Muhammad Umer, Emily FoxLLM Alignment Direct Preference Optimization
Stanford University, Stanford, California, USA
VISPA: Pluralistic Alignment via Automatic Value Selection and Activation
Jan 19, 2026Shenyan Zheng, Jiayou Zhong, Anudeex Shetty +3LLM Alignment Multi-Objective Language Model Alignment
University of Waterloo · University of Melbourne · Macquarie University +2
Untangling Input Language from Reasoning Language: A Diagnostic Framework for Cross-Lingual Moral Alignment in LLMs
Jan 15, 2026Nan Li, Bo Kang, Tijl De BieMultilingual Language Model Evaluation LLM Alignment
IDLab, Department of Electronics and Information Systems Ghent University, Belgium
Asymptotic Universal Alignment: A New Alignment Framework via Test-Time Scaling
Jan 13, 2026Yang Cai, Weiqiang ZhengLLM Alignment Test-Time Scaling
Yale University

Feb 24, 2026J/K move · Enter open · S save

Pressure Reveals Character: Behavioural Alignment Evaluation at Depth

Nora Petrova, John Burden

Prolific

LLM Alignment LLM Safety Benchmarks LLM Evaluation

Open paper AskPDF arXiv

Evaluating alignment in language models requires testing how they behave under realistic pressure, not just what they claim they would do. While alignment failures increasingly cause real-world harm, comprehensive evaluation frameworks with realistic multi-turn scenarios remain lacking. We introduce an alignment benchmark spanning 904 scenarios across six categories -- Honesty, Safety, Non-Manipulation, Robustness, Corrigibility, and Scheming -- validated as realistic by human raters. Our scenarios place models under conflicting instructions, simulated tool access, and multi-turn escalation to reveal behavioural tendencies that single-turn evaluations miss. Evaluating 24 frontier models using LLM judges validated against human annotations, we find that even top-performing models exhibit gaps in specific categories, while the majority of models show consistent weaknesses across the board. Factor analysis reveals that alignment behaves as a unified construct (analogous to the g-factor in cognitive research) with models scoring high on one category tending to score high on others. We publicly release the benchmark and an interactive leaderboard to support ongoing evaluation, with plans to expand scenarios in areas where we observe persistent weaknesses and to add new models as they are released.