{"total":95,"rows":[{"id":"2412.14093","title":"Alignment faking in large language models","created":"2024-12-18","cited_by_count":360,"area":"alignment"},{"id":"2511.18397","title":"Natural Emergent Misalignment from Reward Hacking in Production RL","created":"2025-11-23","cited_by_count":136,"area":"alignment"},{"id":"2509.23909","title":"EditScore: Unlocking Online RL for Image Editing via High-Fidelity Reward Modeling","created":"2025-09-28","cited_by_count":64,"area":"alignment"},{"id":"2503.20823","title":"Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy","created":"2025-03-26","cited_by_count":44,"area":"alignment"},{"id":"2604.13602","title":"Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges","created":"2026-04-15","cited_by_count":33,"area":"alignment"},{"id":"2509.11543","title":"UI-S1: Advancing GUI Automation via Semi-online Reinforcement Learning","created":"2025-09-15","cited_by_count":32,"area":"alignment"},{"id":"2511.23002","title":"JarvisEvo: Towards a Self-Evolving Photo Editing Agent with Synergistic Editor-Evaluator Optimization","created":"2025-11-28","cited_by_count":31,"area":"alignment"},{"id":"2509.23738","title":"GUI-Shepherd: Reliable Process Reward and Verification for Long-Sequence GUI Tasks","created":"2025-09-28","cited_by_count":26,"area":"alignment"},{"id":"2411.12882","title":"ProSec: Fortifying Code LLMs with Proactive Security Alignment","created":"2024-11-19","cited_by_count":25,"area":"alignment"},{"id":"2507.04250","title":"Just Enough Shifts: Mitigating Over-Refusal in Aligned Language Models with Targeted Representation Fine-Tuning","created":"2025-07-06","cited_by_count":24,"area":"alignment"},{"id":"2508.03905","title":"Sotopia-RL: Reward Design for Social Intelligence","created":"2025-08-05","cited_by_count":24,"area":"alignment"},{"id":"2503.10990","title":"Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium","created":"2025-03-14","cited_by_count":23,"area":"alignment"},{"id":"2512.16899","title":"Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image","created":"2025-12-18","cited_by_count":20,"area":"alignment"},{"id":"2505.20259","title":"Lifelong Safety Alignment for Language Models","created":"2025-05-26","cited_by_count":15,"area":"alignment"},{"id":"2507.19672","title":"Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges","created":"2025-07-25","cited_by_count":15,"area":"alignment"},{"id":"2508.06249","title":"In-Training Defenses against Emergent Misalignment in Language Models","created":"2025-08-08","cited_by_count":15,"area":"alignment"},{"id":"2504.15804","title":"Insights from Verification: Training a Verilog Generation LLM with Reinforcement Learning with Testbench Feedback","created":"2025-04-22","cited_by_count":14,"area":"alignment"},{"id":"2506.05339","title":"Flattery, Fluff, and Fog: Diagnosing and Mitigating Idiosyncratic Biases in Preference Models","created":"2025-06-05","cited_by_count":13,"area":"alignment"},{"id":"2506.21463","title":"Aligning Spoken Dialogue Models from User Interactions","created":"2025-06-26","cited_by_count":11,"area":"alignment"},{"id":"2509.19333","title":"Pluralistic Off-policy Evaluation and Alignment","created":"2025-09-15","cited_by_count":11,"area":"alignment"},{"id":"2503.18665","title":"Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark","created":"2025-03-24","cited_by_count":9,"area":"alignment"},{"id":"2504.10886","title":"Exploring Persona-dependent LLM Alignment for the Moral Machine Experiment","created":"2025-04-15","cited_by_count":9,"area":"alignment"},{"id":"2506.10054","title":"Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs","created":"2025-06-11","cited_by_count":9,"area":"alignment"},{"id":"2601.03468","title":"Understanding Reward Hacking in Text-to-Image Reinforcement Learning","created":"2026-01-06","cited_by_count":9,"area":"alignment"},{"id":"2603.12247","title":"Trust Your Critic: Robust Reward Modeling and Reinforcement Learning for Faithful Image Editing and Generation","created":"2026-03-12","cited_by_count":9,"area":"alignment"},{"id":"2503.17338","title":"Capturing Individual Human Preferences with Reward Features","created":"2025-03-21","cited_by_count":8,"area":"alignment"},{"id":"2412.00624","title":"VideoSAVi: Self-Aligned Video Language Models without Human Supervision","created":"2024-12-01","cited_by_count":7,"area":"alignment"},{"id":"2504.12083","title":"Self-alignment of Large Video Language Models with Refined Regularized Preference Optimization","created":"2025-04-16","cited_by_count":7,"area":"alignment"},{"id":"2511.19517","title":"Automating Deception: Scalable Multi-Turn LLM Jailbreaks","created":"2025-11-24","cited_by_count":7,"area":"alignment"},{"id":"2512.23461","title":"Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance","created":"2025-12-29","cited_by_count":7,"area":"alignment"},{"id":"2502.14619","title":"Reward Models Identify Consistency, Not Causality","created":"2025-02-20","cited_by_count":6,"area":"alignment"},{"id":"2505.14185","title":"Safety Subspaces are Not Linearly Distinct: A Fine-Tuning Case Study","created":"2025-05-20","cited_by_count":6,"area":"alignment"},{"id":"2510.01167","title":"Simultaneous Multi-objective Alignment Across Verifiable and Non-verifiable Rewards","created":"2025-10-01","cited_by_count":6,"area":"alignment"},{"id":"2510.08697","title":"BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution","created":"2025-10-09","cited_by_count":6,"area":"alignment"},{"id":"2603.03192","title":"MoD-DPO: Towards Mitigating Cross-modal Hallucinations in Omni LLMs using Modality Decoupled Preference Optimization","created":"2026-03-03","cited_by_count":6,"area":"alignment"},{"id":"2603.20957","title":"Alignment Whack-a-Mole : Finetuning Activates Verbatim Recall of Copyrighted Books in Large Language Models","created":"2026-03-21","cited_by_count":6,"area":"alignment"},{"id":"2502.03397","title":"SPRI: Aligning Large Language Models with Context-Situated Principles","created":"2025-02-05","cited_by_count":5,"area":"alignment"},{"id":"2506.02553","title":"Response-Level Rewards Are All You Need for Online Reinforcement Learning in LLMs: A Mathematical Perspective","created":"2025-06-03","cited_by_count":5,"area":"alignment"},{"id":"2603.15117","title":"MMKU-Bench: A Multimodal Update Benchmark for Diverse Visual Knowledge","created":"2026-03-16","cited_by_count":5,"area":"alignment"},{"id":"2508.09473","title":"NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs","created":"2025-08-13","cited_by_count":4,"area":"alignment"},{"id":"2510.02850","title":"Reward Model Routing in Alignment","created":"2025-10-03","cited_by_count":4,"area":"alignment"},{"id":"2511.19458","title":"Personalized Reward Modeling for Text-to-Image Generation","created":"2025-11-21","cited_by_count":4,"area":"alignment"},{"id":"2602.13891","title":"GSRM: Generative Speech Reward Model for Speech RLHF","created":"2026-02-14","cited_by_count":4,"area":"alignment"},{"id":"2603.15684","title":"State-Dependent Safety Failures in Multi-Turn Language Model Interaction","created":"2026-03-15","cited_by_count":4,"area":"alignment"},{"id":"2604.05279","title":"Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition","created":"2026-04-07","cited_by_count":4,"area":"alignment"},{"id":"2503.23777","title":"CONGRAD:Conflicting Gradient Filtering for Multilingual Preference Alignment","created":"2025-03-31","cited_by_count":3,"area":"alignment"},{"id":"2506.01495","title":"C-VARC: A Large-Scale Chinese Value Rule Corpus for Value Alignment of Large Language Models","created":"2025-06-02","cited_by_count":3,"area":"alignment"},{"id":"2506.02460","title":"MidPO: Dual Preference Optimization for Safety and Helpfulness in Large Language Models via a Mixture of Experts Framework","created":"2025-06-03","cited_by_count":3,"area":"alignment"},{"id":"2508.16921","title":"Being Kind Isn't Always Being Safe: Diagnosing Affective Hallucination in LLMs","created":"2025-08-23","cited_by_count":3,"area":"alignment"},{"id":"2511.13765","title":"PROF: An LLM-based Reward Code Preference Optimization Framework for Offline Imitation Learning","created":"2025-11-14","cited_by_count":3,"area":"alignment"},{"id":"2601.04033","title":"Thinking with Frames: Generative Video Distortion Evaluation via Frame Reward Model","created":"2026-01-07","cited_by_count":3,"area":"alignment"},{"id":"2602.20670","title":"CAMEL: Confidence-Gated Reflection for Reward Modeling","created":"2026-02-24","cited_by_count":3,"area":"alignment"},{"id":"2605.18592","title":"AMARIS: A Memory-Augmented Rubric Improvement System for Rubric-Based Reinforcement Learning","created":"2026-05-18","cited_by_count":3,"area":"alignment"},{"id":"2509.23564","title":"Clean First, Align Later: Benchmarking Preference Data Cleaning for Reliable LLM Alignment","created":"2025-09-28","cited_by_count":2,"area":"alignment"},{"id":"2510.03231","title":"Reward Models are Metrics in a Trench Coat","created":"2025-10-03","cited_by_count":2,"area":"alignment"},{"id":"2510.07315","title":"SWE-IF: Aligning Code Evaluation with Human Preference","created":"2025-10-08","cited_by_count":2,"area":"alignment"},{"id":"2601.14032","title":"RM-Distiller: Exploiting Generative LLM for Reward Model Distillation","created":"2026-01-20","cited_by_count":2,"area":"alignment"},{"id":"2602.12180","title":"How Sampling Shapes LLM Alignment: From One-Shot Optima to Iterative Dynamics","created":"2026-02-12","cited_by_count":2,"area":"alignment"},{"id":"2605.26158","title":"Furina: Fragmented Uncertainty-Driven Refusal Instability Attack","created":"2026-05-24","cited_by_count":2,"area":"alignment"},{"id":"2607.28609","title":"OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models","created":"2026-07-30","cited_by_count":2,"area":"alignment"},{"id":"2503.22480","title":"Probabilistic Uncertain Reward Model","created":"2025-03-28","cited_by_count":1,"area":"alignment"},{"id":"2506.12382","title":"Exploring the Secondary Risks of Large Language Models","created":"2025-06-14","cited_by_count":1,"area":"alignment"},{"id":"2506.13702","title":"Value-Free Policy Optimization via Reward Partitioning","created":"2025-06-16","cited_by_count":1,"area":"alignment"},{"id":"2509.25562","title":"IRIS: Intrinsic Reward Image Synthesis","created":"2025-09-29","cited_by_count":1,"area":"alignment"},{"id":"2602.12660","title":"Learning Ordinal Probabilistic Reward from Preferences","created":"2026-02-13","cited_by_count":1,"area":"alignment"},{"id":"2602.16802","title":"References Improve LLM Alignment in Non-Verifiable Domains","created":"2026-02-18","cited_by_count":1,"area":"alignment"},{"id":"2603.18528","title":"Correlation-Weighted Multi-Reward Optimization for Compositional Generation","created":"2026-03-19","cited_by_count":1,"area":"alignment"},{"id":"2604.17299","title":"Cat-DPO: Category-Adaptive Safety Alignment","created":"2026-04-19","cited_by_count":1,"area":"alignment"},{"id":"2606.10217","title":"Alignment Defends LLMs from Property Inference Attacks","created":"2026-06-08","cited_by_count":1,"area":"alignment"},{"id":"2606.25442","title":"PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models","created":"2026-06-24","cited_by_count":1,"area":"alignment"},{"id":"2503.24150","title":"Learning a Canonical Basis of Human Preferences from Binary Ratings","created":"2025-03-31","cited_by_count":0,"area":"alignment"},{"id":"2506.03557","title":"BPO: Revisiting Preference Modeling in Direct Preference Optimization","created":"2025-06-04","cited_by_count":0,"area":"alignment"},{"id":"2508.05977","title":"LinguaFluid: Language Guided Fluid Control via Semantic Rewards in Reinforcement Learning","created":"2025-08-08","cited_by_count":0,"area":"alignment"},{"id":"2509.08022","title":"DiverValue-Bench: A Benchmark and Fine-Tuning Framework for Aligning Large Language Models with Diverse Human Values","created":"2025-09-09","cited_by_count":0,"area":"alignment"},{"id":"2510.09004","title":"Decoupling Safety into Orthogonal Subspace: Cost-Efficient and Performance-Preserving Alignment for Large Language Models","created":"2025-10-10","cited_by_count":0,"area":"alignment"},{"id":"2602.22745","title":"SPATIALALIGN: Aligning Dynamic Spatial Relationships in Video Generation","created":"2026-02-26","cited_by_count":0,"area":"alignment"},{"id":"2603.06424","title":"From Prompting to Preference Optimization: A Comparative Study of LLM-based Automated Essay Scoring","created":"2026-03-06","cited_by_count":0,"area":"alignment"},{"id":"2603.09246","title":"Reasoning-Oriented Programming: Chaining Semantic Gadgets to Jailbreak Large Vision Language Models","created":"2026-03-10","cited_by_count":0,"area":"alignment"},{"id":"2603.10494","title":"Coverage-Controlled Preference Mining from Noisy Claim Verification for Evidence-Grounded Generation","created":"2026-03-11","cited_by_count":0,"area":"alignment"},{"id":"2604.02574","title":"Understanding the Effects of Safety Unalignment on Large Language Models","created":"2026-04-02","cited_by_count":0,"area":"alignment"},{"id":"2604.07754","title":"The Art of (Mis)alignment: How Fine-Tuning Methods Effectively Misalign and Realign LLMs in Post-Training","created":"2026-04-09","cited_by_count":0,"area":"alignment"},{"id":"2605.17971","title":"Babel: Jailbreaking Safety Attention via Obfuscation Distribution Optimized Sampling","created":"2026-05-18","cited_by_count":0,"area":"alignment"},{"id":"2605.21883","title":"Token-weighted Direct Preference Optimization with Attention","created":"2026-05-21","cited_by_count":0,"area":"alignment"},{"id":"2605.28030","title":"SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance-Diversity Data Selection","created":"2026-05-27","cited_by_count":0,"area":"alignment"},{"id":"2605.28826","title":"From Context Shift to Stylistic Collapse: Why Training Objectives Matter More Than Scale","created":"2026-04-08","cited_by_count":0,"area":"alignment"},{"id":"2606.29097","title":"TrafficAlign: Aligning Large Language Models for Traffic Scenario Generation","created":"2026-06-27","cited_by_count":0,"area":"alignment"},{"id":"2607.07993","title":"Hallucination Self-Play: Bootstrapping Reinforced Detector via Evolved Generator","created":"2026-07-08","cited_by_count":0,"area":"alignment"},{"id":"2607.18258","title":"S2T-RLHF: Hierarchical Credit Assignment for Stable Preference-Based RLHF","created":"2026-05-15","cited_by_count":0,"area":"alignment"},{"id":"2607.20485","title":"Expectation Alignment of Language Models for Real-World User Expectations","created":"2026-06-02","cited_by_count":0,"area":"alignment"},{"id":"2607.23153","title":"In-Context Learning as Implicit Policy Gradient","created":"2026-07-25","cited_by_count":0,"area":"alignment"},{"id":"2608.19748","title":"Truncate Bad, Upweight Good: BoN-Style Distillation via Rank-Based Classification","created":"2026-08-20","cited_by_count":0,"area":"alignment"},{"id":"2609.00879","title":"Towards reliable multimodal disaster severity assessment through preference optimization and explainable vision-language reasoning","created":"2026-09-01","cited_by_count":0,"area":"alignment"},{"id":"2609.03806","title":"SVG-Score: Human-Aligned Evaluation of Text-to-SVG Generation","created":"2026-09-03","cited_by_count":0,"area":"alignment"},{"id":"2609.33989","title":"RewardExplainer: Learning Reward Model Explanations from Counterfactual Preference Feedback","created":"2026-09-27","cited_by_count":0,"area":"alignment"},{"id":"2610.12707","title":"Unsupervised Multimodal Reward Modeling","created":"2026-10-08","cited_by_count":0,"area":"alignment"}],"next_cursor":null}