Preference Learning

Ask about this topic

Overview

Preference learning studies how to infer what people value from pairwise comparisons, rankings, choices, behavior, and other feedback. Common approaches include Bradley–Terry and Gaussian-process models, active query selection, inverse reinforcement learning, and reward-model training with methods such as direct preference optimization. These techniques support personalized recommendations, robot and agent behavior, and multi-objective decisions where no single fixed reward captures every user’s priorities. A central challenge is that preferences can be noisy, context-dependent, and heterogeneous, so researchers must balance efficient data collection and accurate prediction against robust generalization and faithful representation of differing values.

Latest papers

CardsList

149 papers19 papers in the last four weeks, up 36% on the four weeks before. 0.1% of all new papers.

Sep 2, 2026

Toward Collective-Centric Evaluation of Preference Inference for Participatory Democracy

To scale up collective decision-making, participatory democracy platforms such as Polis and Remesh enable online deliberation among thousands of participants. However, at this scale, participants cannot review every opinion submitted by others, producing highly sparse voting data that misrepresent patterns of consensus, conflict, and minority support. Platforms therefore increasingly rely on Preference Inference (PI) models to predict missing votes. Yet this automation is not neutral: inferred preferences can artificially amplify, suppress, or reorder existing patterns of support, ultimately reshaping how the outcomes of a deliberation are interpreted. More generally, we lack a systematic understanding of how existing PI methods affect the collective preference landscape. To address this gap, we benchmark several existing PI approaches in this context. Moving beyond conventional user-centric evaluations centered on the accuracy of individual predictions, we introduce a collective-centric evaluation framework that measures whether inferred votes preserve salient properties of the broader preference landscape. We further contribute the largest multilingual dataset of its kind: four consultations spanning over 90k participants, 1M votes, and 22 languages. Our experiments show that models with comparable predictive accuracy can differ substantially in the degree to which they preserve the collective structure. These results demonstrate that accuracy alone is insufficient for evaluating PI in democratic settings. By contributing a novel comprehensive and collective-centric evaluation benchmark for the task of PI, this work aims to support the development of AI systems that scale deliberation without compromising the integrity of its democratic outcomes.

Pierre-Antoine Lequeu, Salim Hafid, Paul Lerner +6

Sorbonne Université, CNRS, ISIR, Paris, France · Sciences Po, médialab, Paris, France · Sorbonne Université, STIH/CERES, Paris, France

User Preference ModelingPreference Learning
AskPDF
Sep 2, 2026

Eliciting ESG Preferences for Reinforcement Learning-Based Portfolio Optimization

Modern portfolio management increasingly demands a balance between traditional risk-adjusted returns and strict Environmental, Social, and Governance (ESG) mandates. Current Reinforcement Learning (RL) approaches typically optimize for a single ESG provider, neglecting the significant divergence in rating methodologies across the industry and the unintuitive nature of manually weighting conflicting objectives. This paper addresses these limitations by formulating ESG-aware portfolio optimization as a Multi-Objective Reinforcement Learning (MORL) problem that simultaneously incorporates ratings from three distinct ESG agencies. To bridge the gap between high-dimensional algorithmic trade-offs and human decision-making, we integrate a Preference Elicitation framework using Gaussian Processes. This system enables practitioners to infer their latent utility functions through intuitive pairwise comparisons of candidate portfolios based on their Sharpe ratios and aggregate ESG scores. We systematically evaluate our framework by employing Large Language Model (LLM) personas to simulate Portfolio Managers operating under varied regional contexts. Empirical results using historical market data reveal that regional backgrounds fundamentally shift the derived preference weights. For instance, European-based personas tend to prioritize ESG alignment over financial returns, while Texas-based personas favor risk-adjusted performance. This work offers a highly adaptable framework that successfully aligns multi-objective algorithmic trading with diverse, real-world human sustainability preferences.

Giovanni Dispoto, Marcello Restelli, Carmine Ventre

Politecnico di Milano · King’s College London

Portfolio OptimizationFinanceMulti-Objective Reinforcement LearningPreference ElicitationPreference Learning
AskPDF
Sep 2, 2026

Propose to Learn, Learn to Propose: Evaluability-Aware Assistance under Bounded Rationality

AI assistants often collaborate by proposing candidate edits, plans, or designs that users evaluate before adoption. Existing assistance methods focus on proposal quality or user-goal inference, often assuming that the user can reliably evaluate any proposal, which can fail in practice because of bounded rationality. We study evaluability-aware proposal planning, where proposals serve both as task interventions and as probes for learning latent preferences and evaluation constraints, where the resulting belief updates then guide later proposals. We formalise this setting as ProSE, a hidden-parameter sequential assistance problem, and instantiate it with a KL-regularised bounded-rational binary response model in which acceptance trades off value gain against a distance-dependent evaluability penalty. Analysing the planning consequence of this likelihood reveals that likely accepted proposals and informative probes need not coincide, which explains why planners that only pursue acceptance systematically underperform. We operationalise ProSE with ProSE-Plan, a depth-2 Bayes-adaptive planner that scores proposals by possible responses and response-induced posterior beliefs. In controlled graph simulations, ProSE-Plan improves over evaluability-unaware and myopic baselines when evaluation cost is the bottleneck, and a probe-commit ablation confirms that our approach selects informative proposals that simpler methods miss. Our results thus identify user evaluability as a planning-relevant dimension of AI assistance, complementary to generation quality and preference inference.

Yifan Zhu, Sammie Katt, Samuel Kaski

1ELLIS Institute Finland · Department of Computer Science, Aalto University, Finland · Department of Computer Science, University of Manchester, United Kingdom

Active Preference LearningHuman-AI InteractionBounded RationalityPreference Learning
AskPDF
Aug 31, 2026

Low-Resource Preference Adaptation of LLMs via Activation-Based Label Propagation

Adapting large language models to user-specific preferences is often constrained by the cost of human annotation, making preference optimisation impractical in low-resource settings where preferences cannot be reliably labelled by LLMs themselves, e.g., due to cultural, subjective, or personalised contexts. In this paper, we investigate how language models encode preference information in their intermediate representations, finding that activations from chosen and rejected responses form distinct clusters across layers, even in pretrained models. Strikingly, this structure is strengthened by alignment on canonical datasets but erased when the target preferences differ from those the model was aligned on, suggesting aligned LLMs are poor judges for non-mainstream populations. Exploiting this structure, we propose training a lightweight linear probe on a few labelled preference pairs (≤\leq500) and using it to annotate large unlabelled datasets (50K+) for downstream preference optimisation. We systematically evaluate this approach across different datasets, preference optimisation methods and model scales and find that our method consistently outperforms direct training given the same annotation budget, and remains competitive against baselines trained on 50−100×50-100\times more labelled data in the majority of our settings. Code is available at https://github.com/alessioGalatolo/activ-pref-probe.

Alessio Galatolo, Meriem Beloucif

Uppsala University

Pairwise Preference LearningLLM AlignmentUser Preference ModelingPreference Learning
AskPDF
Aug 31, 2026

PLC-DPO: Posterior Label Correction in Noisy and Ambiguous Preference Optimization

Direct Preference Optimization (DPO) simplifies alignment through pairwise comparisons but assumes all observed preferences are reliable. Real data often violates this assumption, leading to reversed, weak, or ambiguous labels that cause harmful policy updates. To address this, we propose Posterior Label Correction DPO (PLC-DPO) to robustly optimize preferences by routing each pair's training signal as a clean, flip, or tie case. The key idea is to use the calibrated policy-reference margin as online evidence to take appropriate correction actions. This reframes noisy preference learning as actively correcting supervision direction and strength rather than merely filtering suspicious examples. Across 57 dataset-model-benchmark cells, PLC-DPO obtains the best mean win rate against DPO (60.5 vs. 55.5 for the next-best method). Injected-noise and tie stress tests, human disagreement analysis, and self-confirmation diagnostics further show that the routing remains stable and distinguishes flipped from weakly directional pairs.

Boryeong Cho, Sumyeong Ahn, Se-Young Yun

KAIST AI · KENTECH

Pairwise Preference LearningDirect Preference OptimizationPreference AlignmentPreference Learning
AskPDF
Aug 27, 2026

Emotional Preferences as Goal-Priority Regulation

A core question in decision-making for agents is whether the relative priorities of competing lower-level objectives can be determined by emotional preferences autonomously generated by higher-level goals, rather than being externally prespecified. Under changing external environments and evolving internal states, emotions play an important functional role in regulating the relative priorities of competing goals. Inspired by the goal-directed theory of emotion, this paper studies how such preference regulation can be computationally realized through reinforcement learning. We first propose a conception of emergent emotional preference: a high-level goal autonomously induces state-dependent preferences over competing lower-level objectives. This conception is built upon a framework consisting of a multi-objective reinforcement learning inner controller and an outer preference generator. The inner controller provides a repertoire of preference-conditioned goal-directed behaviors, while the outer preference generator learns a mapping from the current state to objective preferences through reinforcement learning on a high-level goal. We operationalize emotional preference as a state-dependent regulation of relative goal priorities that emerges through optimization. Furthermore, we characterize the policy space induced by preference regulation and derive an upper bound on the optimality gap in terms of the representation error of the inner behavioral repertoire. We show that the gap vanishes when the optimal policy can be represented by the available preference-conditioned policies. Experiments in self-constructed multi-objective exploration environments show that the learned preference function exhibits contextual priority switching, graded trade-offs, and temporal persistence, and outperforms the evaluated fixed-preference and handcrafted-preference strategies.

Shiqi Liu, Yihua Tan, Hu Fu +1

School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Luoyu Street, Wuhan, 430074, Hubei, China

Reinforcement LearningPreference LearningMulti-Objective Reinforcement Learning
AskPDF
Aug 27, 2026

Disentangling Optimization Scale from Preference Scale in DPO

Direct Preference Optimization (DPO) is a widely used objective for aligning language models from preference data, with the coefficient ββ commonly interpreted as controlling the KL constraint to a reference policy. We show that ββ entangles two distinct roles: it governs the effective inverse preference-noise scale and simultaneously rescales the optimization dynamics, coupling this scale with the effective step size. As a consequence, at a fixed learning rate the achieved policy deviation is non-monotone in ββ: it vanishes in a dead zone at small ββ, reaches a peak at an intermediate value, and decreases again for larger ββ. Moreover, standard DPO loss values are not comparable across ββ: runs with nearly identical loss curves can differ several-fold in KL divergence from the reference model. This entanglement obscures the role of ββ, increases sensitivity to hyperparameter choices, and complicates learning-rate scheduling. We propose a centered-softplus reformulation that is argmin-equivalent to DPO for β>0β>0, while making the inverse preference-noise-scale and learning-rate effects explicit and independently tunable. The normalized centered-softplus objective also admits a continuous β→0β\to0 endpoint that reduces to a linear preference-margin objective.

Ivan Kruzhilov

Direct Preference OptimizationPreference LearningDeep Learning OptimizationRL for Language Models
AskPDF
Aug 27, 2026

Instruction Quality Matters: Refining Instructions for Effective Preference Learning

Preference learning optimizes models using response pairs, yet the informativeness of these pairs is fundamentally shaped by the instructions from which they are generated. We identify instruction quality as a hidden bottleneck in preference learning: low-quality or ambiguous instructions restrict the response-quality distribution, limiting strong chosen responses and weakening preference signals. Through Best- and Worst-of-N analyses, we show that instruction quality constrains both the ceiling and floor of sampled response quality. Motivated by this observation, we introduce an instruction-refinement pipeline that selects weak instructions using reward signals and revises them with rubric-guided LLM feedback, improving preference data without discarding examples. Across offline and online preference learning settings, experiments on multiple models and benchmarks show broad alignment improvements over original data and alternative data-improvement strategies. Further analyses indicate that instruction refinement raises achievable response quality and complements response-centric preference data curation. Overall, instruction quality emerges as a key factor governing how informative preference signals are formed for LLM alignment. Code is available at: https://github.com/01choco/instruction-refinement/

Seohyeong Lee, Hwaran Lee, Buru Chang

Sogang University · Korea University

Preference LearningLLM AlignmentPreference Alignment
AskPDF
Aug 26, 2026

Beyond Pairwise Feedback: Listwise Vision-Language Supervision for Preference-Based Reward Learning

Vision-language models (VLMs) have emerged as a powerful source of supervision for reinforcement learning, enabling agents to leverage rich semantic knowledge during training. Inspired by the success of preference-based reward learning (PbRL) in reinforcement learning from human feedback (RLHF), vision-language model generated image-based preferences provide an effective source for learning reward functions. This can be done by visually comparing two outcomes through the Bradley-Terry (BT) model. However, this pairwise formulation utilizes only two observations at a time, despite VLMs being capable of ranking multiple candidates. The Plackett-Luce (PL) formulation can shape a reward model with listwise rankings as opposed to pairwise preferences, allowing for a more suited use of a VLM based ranking. In this work, to our knowledge, we introduce the first framework that combines VLM-generated preferences with the Plackett-Luce model for reward learning. We evaluate our approach on Meta-World manipulation tasks and show that Plackett-Luce (PL) reward models can train robotic policies from VLM-generated rankings as effectively as pairwise Bradley-Terry, KK-wise Bradley-Terry, and RL-VLM-F baselines. Across all environments, at least one PL ranking size (K∈{3,4,5}K \in \{3,4,5\}) consistently performs with or outperforms other methods in mean success rate. Unlike pairwise methods, which are restricted to K=2K=2, PL supports different ranking sizes and can therefore be adapted to the environment and desired feedback format. Our best PL configuration achieves an 86% mean final success rate and matches the Oracle baseline on Drawer Open. Overall, these results demonstrate that listwise VLM preference supervision is a competitive and flexible approach to reward learning for reinforcement learning.

Srivalli Katkuri, Maxwell Kawada, Juan Wachs

School of Mechanical Engineering Purdue University West Lafayette, USA · Weldon School of Biomedical Engineering Purdue University West Lafayette, USA · Edwardson School of Industrial Engineering Purdue University West Lafayette, USA

Preference-Based RLPreference LearningMultimodal Reward ModelingVLMs for Robotics
AskPDF
Aug 14, 2026

AI Research Preference Models

AI research agents (AIRA) can now carry machine learning experiments from proposal through implementation and evaluation. Yet progress on frontier tasks is throttled by the cost of evaluations that can consume days of GPU time. When an agent can propose far more candidates than it can afford to run, progress depends on its research preference: how it allocates a fixed execution budget across many candidates. We introduce AI Research Preference Models (RPMs) that predict which candidate solution is most promising, without paying the cost of running them all. We build RPMs from frozen pretrained language models in two variants: an inference-only model that reasons over candidate plans, code, and previously executed solutions, and an agentic model that additionally runs small-scale pilot experiments. Integrated into the AIRA-dojo research agent and evaluated on the machine learning research benchmark AIRS-Bench, the two variants increase the average normalized score from 0.684 to 0.711 and 0.729, respectively. Both reach the unguided agent's 24-hour performance in roughly 15 hours, using less than two-thirds of its execution budget, and together yield new state-of-the-art results on two AIRS-Bench tasks.

Thomas Simon Foster, Bassel Al Omari, Tingchen Fu +30

FAIR at Meta · University of Oxford · University College London

AI Agents for Scientific DiscoveryPreference Learning
AskPDF
Aug 13, 2026

HumanTracker: Towards Comprehensive and Human-Aligned Motion Tracking Benchmark

Humanoid motion tracking is central to teleoperation and whole-body imitation, yet evaluation often disagrees with what people perceive in videos. Kinematic errors average per-frame pose differences but miss the physical artifacts that matter most, particularly unstable support and incorrect contacts such as foot skating and mistimed touch-downs. Meanwhile, widely used test suites are small and lack the diversity needed to stress contact-rich, long-horizon behaviors. We introduce HumanTracker to make humanoid tracking evaluation both perceptually aligned and scalable. The HumanTracker benchmark contains approximately 153 hours of optical motion trajectories from multiple professional performers, organized into four motion families with text labels for fine-grained diagnosis. We further propose HumanScore, a preference-aligned metric trained on 12K motion pairs containing 24K motions. Across representative state-of-the-art trackers, HumanScore better predicts human preferences and reveals contact and stability failures that kinematic metrics often miss.

Dairu Liu, Zekun Qi, Jiayu Zeng +11

1Nankai University · 3Galbot · 2Tsinghua University +3

Human-in-the-Loop EvaluationHumanoid Motion TrackingPreference LearningHuman Motion Understanding
AskPDF
Aug 4, 2026

Learning Sexism Detection Using Multi-Agent Perspectivist Preference Optimization

When people label text for sexism, they often disagree, and not because some of them are wrong: they genuinely perceive sexism differently. Most NLP systems discard this disagreement by collapsing it into a majority vote. We propose the Multi-Agent Perspectivist Preference Optimization (MAP-PO) framework to keep these different perspectives. On the EXIST 2024 dataset of labeled English and Spanish tweets, we first cluster annotators by their labeling behavior rather than their demographic attributes. We then fine-tune one Large Language Model agent per cluster to reproduce that cluster's annotation behavior, and coordinate the agents with preference optimization that combines individual and team-level rewards. We evaluate MAP-PO in four settings defined by two languages and two backbone language models, asking whether each agent reproduces the annotations of its own cluster and whether the agents together reproduce the majority label. Two findings hold in all four settings. First, without fine-tuning the agents behave almost identically, so cluster-specific training is necessary. Second, we show that training each agent only on the labels of its own cluster pushes the agents far beyond the clusters they should represent, while adding a shared team-level training signal consistently keeps each agent calibrated to its cluster.

Hadi Mohammadi, Tina Shahedi, Robert A. Bagheri +2

Department of Methodology and Statistics, Utrecht University, The Netherlands · Department of Information and Computing Sciences, Utrecht University, The Netherlands · Field Technology Innovations (FTI) & Robotics, Wageningen University & Research, The Netherlands

Annotator DisagreementPreference OptimizationMulti-Agent LLMsPreference Learning
AskPDF
Jul 30, 2026

Learning Color Grading, No Photo Sharing: Federated Aesthetic Preference Learning for Personalized Image Enhancement

Personalized image enhancement should reflect individual aesthetic taste, yet learning such preferences commonly depends on private photos and ratings that are unsuitable for centralized collection. The task must infer preference from sparse, heterogeneous feedback and translate it into natural-looking color transformations on resource-constrained user devices. We introduce FedPAIE, a federated personalized aesthetic image enhancement framework for user-adaptive color grading without centralizing raw photos or ratings. FedPAIE trains a lightweight dual-cue aesthetic scorer, calibrates it into a personalized scorer on a small local support set, and freezes it to guide regularized adaptation of a lightweight CLUT enhancer from unpaired local photographs. Fidelity constraints and an excess-gap penalty regularize scorer-guided adaptation to limit proxy-score over-optimization while preserving content and natural appearance. Training remains lightweight throughout the pipeline: scorer learning updates at most 0.787M parameters, enhancer adaptation updates 0.265M, and inference retains only a 0.293M-parameter personalized enhancer. Experiments on MIT-Adobe FiveK and Flickr-AES demonstrate effective open-world personalization and a favorable balance between user preference and image fidelity. FedPAIE thus connects decentralized preference learning with efficient personalized image transformation without requiring paired user retouches.

Chuanzhi Xu, Ziyuan Tao, Jean Julien KNell +5

The University of Sydney · Macquarie University

Personalized Image Aesthetics AssessmentImage EnhancementPreference LearningFederated Learning
AskPDF
Jul 20, 2026

AdaHome: An Adaptive Smart Home Assistant using Local Small Language Models

Smart home assistants interpret a wide range of user commands, from explicit device control to underspecified and preference dependent requests. While recent systems based on Large Language Models (LLMs) improve this capability, they often rely on heavyweight reasoning pipelines and cloud-based deployment, limiting their efficiency and suitability for resource-constrained environments, and raising privacy concerns. In addition, existing approaches provide limited support for stable long-term personalization. To address these issues, we present AdaHome, an adaptive smart home assistant designed for locally deployed small language models in smart home environments. Rather than applying complex reasoning uniformly, AdaHome introduces an intent-aware planning framework that dynamically routes commands either to straightforward prompt-based or lightweight reasoning-based components. For commands requiring interpretation, we adopt a Chain-of-Draft strategy to enable efficient and stable decision-making. To support personalization, we further propose a preference adaptation mechanism that learns from user feedback over time without requiring prompt augmentation or model retraining. We evaluate AdaHome against representative LLM-based baselines under a unified small model setting. AdaHome achieves substantially higher accuracy on direct commands (86.7%) while reducing latency by up to 3×\times. Furthermore, it maintains competitive performance on ambiguous inputs with lower computational cost. In multi-turn scenarios, AdaHome achieves 88% preference consistency, compared to 52.5% for a prompt augmentation baseline.

Eu Jin Lim, Zhaoxing Li, Sebastian Stein

University of Southampton Southampton, United Kingdom

Adaptive InferenceOn-Device Language Model InferenceSmall Language ModelsLLM RoutingConversational AgentsPreference Learning
AskPDF
Jul 20, 2026

HyCoRec: Hypergraph-Enhanced Multi-Preference Learning for Alleviating Matthew Effect in Conversational Recommendation

The Matthew effect is a notorious issue in Recommender Systems (RSs), i.e., the rich get richer and the poor get poorer, wherein popular items are overexposed while less popular ones are regularly ignored. Most methods examine Matthew effect in static or nearly-static recommendation scenarios. However, the Matthew effect will be increasingly amplified when the user interacts with the system over time. To address these issues, we propose a novel paradigm, Hypergraph-Enhanced Multi-Preference Learning for Alleviating Matthew Effect in Conversational Recommendation (HyCoRec), which aims to alleviate the Matthew effect in conversational recommendation. Concretely, HyCoRec devotes to alleviate the Matthew effect by learning multi-aspect preferences, i.e., item-, entity-, word-, review-, and knowledge-aspect preferences, to effectively generate responses in the conversational task and accurately predict items in the recommendation task when the user chats with the system over time. Extensive experiments conducted on two benchmarks validate that HyCoRec achieves new state-of-the-art performance and the superior of alleviating Matthew effect. Our code is available at https://github.com/zysensmile/HyCoRec.

Yongsen Zheng, Ruilin Xu, Ziliang Chen +4

1Sun Yat-sen University · 2Peng Cheng Laboratory · 5Jinan University +2

Recommender SystemsPopularity Bias in Recommender SystemsPreference Learning
AskPDF
Jul 18, 2026

A Method for Learning Value Systems in Generative AI

Value-aware AI systems require explicit computational representations of human values (groundings) and their aggregation into value systems in order to align their decisions with ours. As such representations are difficult to elicit, value learning seeks to infer them by observing human behaviour. This work addresses the lack of grounded value learning methods in generative AI: existing approaches typically replicate human preferences without awareness of the multidimensional structure of value alignment, or lack principled value system elicitation methods. To address these gaps, we adapt a previously validated value system learning method to the generative AI setting, which, based on pairwise prompt-response preference data, simultaneously learns: i) an implementation of a grounding for a set of values given by a multi-objective reward model, and ii) a value system representation in the form of a weighted linear scalarization of the previous grounding model. To ensure that the learned value systems are based on coherent value representations, our algorithm dynamically prioritizes the grounding learning process. We evaluate the method against baselines and a contemporary method on prompt-response preference datasets. Results show competitive performance and minimal trade-offs against the baselines, while improving explainability.

Andrés Holgado-Sánchez, Holger Billhardt, Sascha Ossowski

1CETINIA, University Rey Juan Carlos

Reward ModelingLLM AlignmentPreference Learning
AskPDF
Jul 17, 2026

Hard Rules, Soft Preferences: Bridging Reasoning, Learning, and Optimization for Personalized Packing Checklist Generation

Packing for air travel is recurring and error-prone: the checklist must be personal and context-aware, yet feasible under safety rules, item dependencies, and luggage limits. Existing packing assistants are template-driven and generic, or recommendation-driven but unconstrained, leaving users to manually patch regulatory and capacity violations. We propose a reasoning-guided learning framework with three stages: (1) a symbolic engine that generates a regulation-aware seed checklist with explicit dependency structure, (2) a two-stage preference learner that estimates inclusion and priority utilities from user add and remove actions while mitigating survivorship bias, and (3) a CP-SAT optimizer that selects a compact, compliant subset. The architecture instantiates a general pattern for constrained personalization, applicable wherever hard feasibility coexists with sparse preference signals. On 604 labeled trip scenarios, comprising 29K inclusion labels and 343K pairwise comparisons, the symbolic engine attains 99.7% recall and 0.96 rubric validity, compared with 0.78 to 0.81 for frontier LLMs. Gradient-boosted trees and LambdaMART reach an AUC-ROC of 0.943 and an NDCG@5 of 0.923. CP-SAT attains 100% constraint satisfaction, compared with 28% for greedy selection and 10% for random selection. Deployment in FlyEnJoy, a production iOS travel app, doubled checklist completions and reduced editing and completion time.

Himel Dev, Madhusudan Basak, Tanmoy Sen +2

Tech LLC · University of Massachusetts Amherst · University of Virginia

Recommender SystemsPreference Learning
AskPDF
Jul 16, 2026

Step-Level Preference Learning for Generative Agents in Social Simulations

Large language model (LLM)-based generative agents simulate human behavior through long-horizon decision-making processes that comprise intermediate steps such as planning, memory retrieval, reflection, and action selection. However, fine-grained human annotations of these intermediate steps remain scarce, and existing agents are not grounded in human preferences over such intermediate decisions. To address this gap, we introduce \method, an interactive simulation interface that enables us to collect step-level human preference supervision over agent decision trajectories, leading to a dataset of 57K fine-grained annotations. We conduct step-level preference learning on open-weight language models using supervised finetuning and direct preference optimization on this data, consistently improving simulation fidelity, coordination, and interaction quality, and inducing more socially effective agent behavior. Our results show that step-level human supervision is an effective training signal for improving both local decision quality and long-horizon agent behavior.

Wenchang Gao, Pingyue Sheng, Lanlan Qiu +7

Shanghai Qi Zhi Institute · Tsinghua University · Xiongan AI Institute +3

Human Preference ModelingHuman Behavior SimulationLLM Agent TrainingLarge Language Model-Based Social SimulationPreference Learning
AskPDF
Jul 13, 2026

On Generalizing Preference-based Reinforcement Learning to Multi-Objective Problems

We study reinforcement learning (RL) from pairwise human-generated trajectory comparisons. We generalize preference-based RL by formalizing a novel setting in which the expert can also label trajectory pairs as incomparable when neither trajectory "dominates" the other. We introduce the learning problem and the desiderata that a model should satisfy, showing that, when multiple objectives are present, no compliant model admits a negative log-likelihood that is convex in the utility difference. Then, we propose a novel Bradley-Terry-inspired rationality model that enables incomparability and infers a multi-dimensional reward function, and discuss its compliance with the desiderata. We upper bound the divergence between the true and learned comparison distributions when a dataset is available, separating the rate of approach to the optimum from a non-vanishing term due to non-convexity. Then, we study how learned parameters affect downstream guarantees, highlighting the impact of invariance on Pareto optimality.

Simone Drago, Marco Mussi, Leonardo Bianconi +1

Politecnico di Milano, Milan, Italy

Pairwise Preference LearningPairwise ComparisonPairwise Preference EvaluationBradley–Terry ModelPreference LearningPreference-Based RL
AskPDF
Jul 6, 2026

Probably Correct Optimal Stable Matching under Two-Sided Uncertainty

We study a sequential learning problem for stable matchings in two-sided markets where preferences on both sides are initially unknown. We focus on a centralized setting where an algorithm matches agents at each time step and receives noisy rewards that reflect the preferences of the matched agents, following a semi-bandit feedback structure. We adopt a pure exploration perspective, aiming to efficiently identify the optimal stable matching with high probability. Our work extends prior results by handling two-sided uncertainty and by exploiting partial preference information. A central ingredient is the notion of pervasive stable matching, which enables the identification of optimal stable matchings under partial preferences. We propose elimination-based algorithms whose stopping criteria exploit the structure of the learned partial preferences, and provide a refined sample-complexity analysis. Beyond pure exploration, we extend our approach to regret minimization and establish regret bounds with respect to the optimal stable matching that avoid dependence on the minimum reward gap Δmin⁡Δ_{\min}.

Andreas Athanasopoulos, Anne-Marie George, Christos Dimitrakakis

University of Neuchâtel, Neuchâtel, Switzerland · University of Oslo, Oslo, Norway

Multi-Armed BanditsPreference Learning
AskPDF
Jul 3, 2026

Unbiased Alignment for Large Language Models with Noisy Preferences

The alignment of large language models with human preferences is commonly achieved through Reinforcement Learning from Human Feedback or Direct Preference Optimization. However, these methods are vulnerable to the significant noise prevalent in real-world preference datasets. To address this critical issue, we present a theoretical framework for unbiased alignment, introducing the Unbiased Reward Model (URM) loss and the Unbiased Direct Preference Optimization (UDPO) loss. By mathematically correcting the distortion induced by preference noise, our novel objectives enable unbiased model training directly from noisy datasets, without requiring clean ground-truth supervision. We provide rigorous theoretical analyses demonstrating that our methods are noise-tolerant, parameter downward compatible, and classification-calibrated. Comprehensive experiments across diverse datasets demonstrate that our approaches outperform state-of-the-art baselines. Code available at: https://github.com/cswjl/unbiased-alignment.

Jialiang Wang, Xianming Liu, Xiong Zhou +2

1Harbin Institute of Technology · 2City University of Hong Kong.

Pairwise Preference LearningLLM AlignmentDirect Preference OptimizationLearning with Noisy LabelsPreference Learning
AskPDF
Jul 2, 2026

SPLC: Social Preference Learning for Crowd Robot Navigation

Offline reinforcement learning (RL) holds significant potential for crowd robot navigation in human-robot coexistence applications. However, the inherent complexity of pedestrian motion renders the design of effective reward functions for promoting socially compliant robot behaviors a persistent challenge. This paper proposes a Social Preference Learning for Crowd Robot Navigation (SPLC) algorithm to eliminate the need for detailed reward design. Its core innovation lies in the introduction of a social preference feedback mechanism to automatically generate preference data through principled preference evaluation criteria. By explicitly accounting for the intricacies of pedestrian dynamics, the pipeline mitigates the reward bias and facilitates the systematic quantification of broad social norms, thereby fostering socially compliant behaviors. Extensive experiments integrating SPLC with offline RL methods demonstrate consistent improvements over state-of-the-art baselines across standard performance metrics. Furthermore, real-world experiments on the TurtleBot4 further validate the effectiveness of SPLC in practical human-robot coexistence settings. Our code and video demos are available at https://github.com/sklus949/SPLC.

Zixuan Chen, Hao Fu, Haiwen Hu +1

Wuhan University of Science and Technology, Wuhan, China · School of Computer Science and Technology, Wuhan University of Science and Technology and also with the Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System, Wuhan 430081, China

Robot NavigationSocial Robot NavigationOffline RLReward Design for RLPreference LearningPreference-Based RL
AskPDF
Jul 2, 2026

Scaling Trends for Lie Detector Oversight in Preference Learning

Deceptive behavior in LLMs is costly to monitor and prevent, motivating approaches such as Scalable Oversight via Lie Detectors (SOLiD) (Cundy & Gleave, 2025), which uses lie detectors to identify responses for review by high-cost labelers. In this paper, we scale SOLiD to larger models and evaluate it in more diverse and realistic preference-learning settings. We find favorable scaling: undetected deception drops from 34% for 1B-parameter models to 14% for 405B-parameter models at a detector true positive rate of 99%, and expensive human labelers can be removed entirely from the fine-tuning phase without a statistically significant increase in deception. However, SOLiD is sensitive to distribution shift between detector training and preference-training data, which can drive detector false positive rates to impractical levels.

Oskar J. Hollinsworth, Ann-Kathrin Dombrowski, Sam Adam-Day +2

Scalable OversightDeception in Language ModelsLLM Safety EvaluationHuman Oversight of AIPreference Learning
AskPDF
Jul 1, 2026

PAPA: Online Personalized Active Preference Alignment

Diffusion models are highly effective at modeling complex data distributions, including images and text. However, in applications like personalized recommender systems, the objective often shifts to modeling specific regions of the distribution that maximize user preferences-initially unknown but gradually uncovered through interactive feedback. This can naturally be framed as a reinforcement learning problem, where the goal is to fine-tune a diffusion model to maximize a reward function based on preferences. However, the main challenge lies in learning a parameterized reward model, which typically requires large-scale preference data-something that is often not feasible in practice. In this work, we introduce Personalized Active Preference Alignment PAPA, a novel method that bypasses the requirement for a parametrized reward model by directly optimizing the diffusion model using real-time user feedback. PAPA enables feedback-efficient preference alignment, drawing inspiration from the variational inference framework. We demonstrate PAPA's effectiveness through extensive experiments and ablation studies across diverse class-conditioned and fine-grained alignment tasks. Additionally, based on theoretical insights, we propose an enhanced fine-tuning strategy, referred to as EPAPA, that requires less computational budget and accelerates the fine-tuning process, further boosting PAPA's suitability for real-world deployment. Our code is made publicly available at https://github.com/NasikNafi/papa.

Anindya Sarkar, Nasik Muhammad Nafi, Isaac Lyngaas +2

Washington University in St. Louis · Oak Ridge National Laboratory

Diffusion Model AlignmentActive Preference LearningDiffusion Model Fine-TuningRL for Diffusion ModelsUser Preference ModelingPreference Learning
AskPDF
Jun 30, 2026

Bridging Local Observation and Global Simulation in Closed-Loop Traffic Modeling

A local-to-global context mismatch arises when autoregressive traffic simulators trained on ego-centric driving logs are deployed in globally observable closed-loop environments. In such logs, the ego vehicle has rich local observations, while surrounding agents are only partially observed due to perception limits and occlusions. As a result, simulators may learn incomplete context--action mappings that remain hidden in log-based training but emerge during closed-loop rollouts, leading to unrealistic behaviors such as abnormal stops, unsafe interactions, and rule violations. We propose CRAFT, a Contextual pReference Alignment Framework for Traffic Simulation, to mitigate this mismatch via self-supervised failure discovery and preference-guided test-time alignment. CRAFT treats the base simulator as a globally observable sandbox, generating diverse what-if rollouts from logged initial states to expose context-induced failures. These failures are grounded with human-aligned driving priors and converted into preference supervision for training a Contextual Preference Evaluator (CPE). At inference time, CPE acts as a plug-in alignment module that scores candidate actions under complete scene context and reweights autoregressive decoding toward globally coherent behaviors. CRAFT mitigates this local-to-global contextual bias, reducing collisions by 31.2% and traffic violations by 33.2% without retraining the base simulator.

Ziyan Wang, Tan Xiang, Peng Chen +1

The University of Hong Kong · Beihang University

Test-Time AdaptationAutonomous Driving SimulationPreference Learning
AskPDF
Jun 24, 2026

MAPL: Multi-Objective Preference Learning for Robot Locomotion

Reward design remains a major bottleneck in reinforcement learning for robot locomotion, where successful policies often depend on carefully tuned, task-specific reward functions. Preference-based reinforcement learning offers an alternative, but existing LLM-based methods typically ask for a single overall judgment between behaviors, making it difficult to capture the multiple competing objectives that underlie high-quality locomotion. We present Multi-Objective AI-Informed Preference Learning (MAPL), a framework that learns locomotion rewards from high-level natural language objectives rather than manually engineered reward equations. MAPL prompts a large language model to compare trajectories independently along semantically meaningful criteria, using generic language descriptions that are terrain-invariant and require little domain expertise. These objective-wise preferences are used to train a multi-head preference scoring model, whose outputs are aggregated to form a scalar reward for policy optimization. Across four quadruped locomotion environments, MAPL trains policies using only LLM-generated preferences and achieves performance comparable to or better than expert-designed rewards, while eliminating task-specific reward engineering.

Xiyue Chen, Muhan Lin, Shuyang Shi +1

Department of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47906, USA · Department of Computer Science, Purdue University, West Lafayette, IN 47906, USA · independent researcher

Multi-Objective Reinforcement LearningReward ModelingRL for Legged LocomotionPolicy LearningReward Design for RLPreference Learning+1
AskPDF
Jun 22, 2026

Towards Spec Learning: Inference-Time Alignment from Preference Pairs

Steering a large language model (LLM) toward a desired behavior typically relies on an iterative process of hand-crafting a prompt based on a careful inspection of the model's responses. This is an involved, brittle, and error-prone process. Preference-based fine-tuning is a more rigorous but often prohibitively expensive solution. We propose spec learning, a framework that relies on a brief user instruction and a small set of preference judgments. These are compiled into specifications in the form of natural-language prompts for an LLM. Specifications condition LLMs at inference time, and no parameter updates to the underlying models are required. We show that the responses generated based on the compiled specifications often outperform direct preference optimization (DPO) on datasets from specialized domains whose preference signal is dense. Unlike opaque weight updates, the resulting specifications are human-readable and double as interpretable and transparent written embodiments of the preference signal that produced them.

Dhriti Krishnan, Tejas Goyal, Jaromir Savelka

Department of Computer Science Carnegie Mellon University

LLM AlignmentPrompt OptimizationInference-Time Language Model AlignmentPreference LearningInference-Time Guidance
AskPDF
Jun 20, 2026

Selective Ensemble Based on Preference-Directed Multi-Objective Bandits

Selective ensemble for modern machine learning systems requires choosing promising model candidates under limited evaluation budgets, while downstream tasks often specify only partial preferences over capabilities such as accuracy, robustness, and reasoning. This setting naturally gives rise to a sequential decision problem under partially specified linear preferences. We formalize it as preference-directed multi-objective bandits (PDMOB), where admissible trade-offs are represented by a polyhedral preference cone. Based on this formulation, we introduce Pareto CC-optimality, which recovers standard Pareto optimality and single-weight scalarization as special cases. We then propose the preference-directed upper confidence bound (PrefUCB) algorithm, which maintains directional confidence intervals to guide exploration. We analyze both indicator-based and gap-weighted regret, and establish instance-dependent logarithmic bounds for both criteria, recovering the optimal logarithmic dependence on the horizon TT in classical special cases. Experiments on large pre-trained model selective ensemble tasks and online asset allocation under institutional mandates validate the efficacy of our method.

Lanjihong Ma, Zhen-Yu Zhang, Masashi Sugiyama +1

Department of Computer Science, Zhejiang Gongshang University, China · Center for Advanced Intelligence Project, RIKEN, Japan · Graduate School of Frontier Sciences, The University of Tokyo, Japan +1

Multi-Armed BanditsEnsemble LearningUCB AlgorithmsCombinatorial Multi-Armed BanditsPareto OptimizationPreference Learning
AskPDF
Jun 17, 2026

Interactive Pareto navigation for deep multi-task learning

In multi-task learning, handling an increasing number of objectives can quickly become challenging, both in terms of the computational resources and the decision maker's capacity to choose appropriate trade-offs. A widely used approach is thus to aggregate the individual losses in a single loss function by a weighted sum. This often fails to capture either the decision maker's preferences as a result of the shape of the Pareto front, or requires multiple adjustments and computations which becomes prohibitively expensive in deep learning applications. To address these issues, we introduce a novel framework, Preference Pareto Exploration (PPE), which enforces the decision maker's preferences while accounting for the geometry of the Pareto set in an interactive exploration process. PPE is based on a predictor-corrector method that performs predictor steps tangential to the manifold of Pareto-optimal solutions, following the decision maker's preference. The subsequent corrector step results in a new trade-off reflecting this preference. To avoid explicit Hessian computations when characterizing the tangent space of the manifold, we employ a Krylov subspace method that relies solely on matrix-vector products. These products can be efficiently obtained via automatic differentiation, ensuring both efficiency and robustness throughout the optimization process. The method's functionality and performance are demonstrated using both toy problems and examples from deep learning.

Augustina C. Amakor, Konstantin Sonntag, Sebastian Peitz

Department of Computer Science, TU Dortmund, Dortmund, Germany · Lamarr Institute for Machine Learning and Artificial Intelligence

Multi-Task LearningPareto OptimizationPreference Learning
AskPDF
Jun 14, 2026

Improving Code-Switching ASR with Code-Mixing Guided Synthetic Speech

Code-switch (CS) Automatic Speech Recognition (ASR) remains challenging due to limited availability of high quality CS text-speech pairs for training. Although synthetic data augmentation via Text-to-speech (TTS) has been explored, existing CS TTS approaches primarily optimise reconstruction fidelity and do not explicitly enforce language-boundary consistency, thereby limiting their effectiveness for CS ASR augmentation. This paper proposes a code-mixing guided preference-learning framework that steers synthetic speech generation toward improved code-switching fidelity using the Code Mixing Index (CMI). Experiments on the SEAME Mandarin-English conversational corpus demonstrate that the proposed method enhances the utility of synthetic data for ASR fine-tuning. Specifically, when fine-tuning Whisper Large, the proposed approach reduces Mixed Error Rate (MER) from 12.1%/17.8% to 8.9%/14.2% on the DevMAN and DevSGE sets, respectively.

Yue Heng Yeo, Haoyang Li, Yizhou Peng +6

College of Computing and Data Science, Nanyang Technological University, Singapore · HLT-COE & CLSP, Johns Hopkins University, USA · Institute for Infocomm Research (I2R), A⋆STAR, Singapore +1

Code-Switching Speech RecognitionSynthetic Data AugmentationTTS SynthesisAutomatic Speech RecognitionSynthetic Data Augmentation for Speech RecognitionPreference Learning+1
AskPDF