Reinforcement learning from human feedback (RLHF) has emerged as a powerful yet sample-inefficient approach for learning reward models from human preferences, making active learning a critical component in synthesizing informative preference queries. However, effective uncertainty quantification required for active learning remains a key challenge for large neural network reward models. In this paper, we introduce PreferenceEKF, a sample-efficient approach that tracks reward model uncertainty by framing active preference learning as a sequential Bayesian filtering problem. Instead of relying on computationally prohibitive posterior inference over the full neural network parameter space, our method performs sequential inference via an extended Kalman filter within a low-dimensional parameter subspace, continuously updating the reward model posterior as new preference queries arrive. Our approach enables scalable sampling of neural network parameters to efficiently compute acquisition functions for active reward learning. Experiments on the D4RL and V-D4RL benchmarks demonstrate that our approach achieves better sample efficiency, runtime, scalability, and calibration compared to other Bayesian deep learning approaches, and the learned reward models lead to competitive offline reinforcement learning policy performance. This highlights the potential of scalable Bayesian methods for preference-based reward modeling in RLHF. Our code is available at https://github.com/yutaizhou/bnn_pref.
Camera-conditioned world models generate interactive videos in which commanded actions should induce the expected scene changes while appearance, geometry, and temporal dynamics remain coherent. Existing rewards assess these requirements separately: geometry-based rewards estimate trajectory execution but cannot judge the visual quality of the executed motion, whereas image-based rewards measure frame quality without capturing action execution or temporal dynamics. We posit that a vision-language model (VLM) offers a shared reasoning space for relating actions to their visual outcomes. However, judging a complete long video against its full action sequence creates a lengthy, noisy context in which short-lived local action evidence can be missed or diluted. We present WorldReward, a VLM-based pairwise preference reward model that unifies action-consistency and visual-quality evaluation for camera-conditioned world models. WorldReward decomposes paired videos into action-aligned chunks, organizes each chunk into structured visual evidence, and aggregates chunk-level decisions by voting into separate video-level action and visual-quality preferences. To train it, we construct a large-scale reasoning-augmented preference dataset using structured judgments generated by a frontier VLM and refined through tool-based agent auditing and targeted human review. We further introduce WorldReward-Bench, a human-annotated benchmark measuring reward-model agreement with human preferences across action consistency, appearance quality, and motion quality. WorldReward achieves the highest agreement on all three dimensions, exceeding GPT-5.5 by 3.42, 1.45, and 3.56 percentage points, respectively. When used for RL post-training of HY-WorldPlay 1.5, it consistently improves both action execution and visual quality across short- to long-term horizons.
Large language models (LLMs) have become ubiquitous tools for code generation and editing. However, development teams often use multiple LLM assistants. Different developers may prefer different models, and individual developers may switch between models across different coding sessions. Because of this, the edits any one model makes are frequently applied to foreign code originally generated by another model. These LLMs are often trained on different datasets, and as a result have different stylistic preferences. Do LLMs behave differently when they edit foreign code originally written by a different LLM with a different coding style? We find that models tend to make more, and often excessive, edits on foreign code. We introduce CROCODIL (Cross-model Code Editing with LLMs), a post-training framework for reducing excessive edits while preserving functional correctness. CROCODIL's similarity reward penalizes large changes, while its execution reward scores build and test success. We use the product of these two rewards to encourage the policy to decrease the edit size without decreasing the edit task success rate. CROCODIL is available at https://github.com/EngineeringSoftware/Crocodil.
Modern vision-language models (VLMs) can directly answer many image-grounded questions, yet they often struggle with complex queries requiring fine-grained visual details or external knowledge. To acquire this missing evidence, agentic VLMs invoke tools such as image cropping, image search, and text search. However, existing training paradigms primarily evaluate tool-use based on final answer correctness, leaving evidence acquisition and utilization insufficiently supervised. This leads to two critical shortcomings: (i) models frequently issue redundant or off-target tool calls that fail to gather necessary evidence, and (ii) even when appropriate tools are called, models often fail to extract the necessary information from the resulting observations. To address these limitations, we introduce the NTEP (Necessary Tool-Evidence Path), a novel annotation scheme that explicitly specifies the essential external evidence and corresponding tool calls for each query. Building upon this, we propose NTEP-R (NTEP Reward), a supervision mechanism ensuring that each tool invocation strictly advances the reasoning process toward the final solution. Specifically, our approach rewards the agent for aligning its pre-call intent with a necessary evidence-seeking goal, and for ensuring the information summarized from the post-call observation aligns with the necessary evidence. Furthermore, we introduce a non-repeated-goal regularizer to penalize redundant calls that revisit satisfied NTEP goals. Extensive evaluations on seven image-grounded benchmarks demonstrate that our 8B-parameter instantiation, NTEP-8B, significantly improves both search-oriented accuracy and tool-use efficiency within a unified three-tool framework. These results highlight the critical value of fine-grained tool-evidence path supervision for training robust agentic VLMs.
Reinforcement learning typically optimizes average reward. For generative policies, the average can hide an important distinction: two policies can achieve the same mean reward while having very different chances of producing a rare but high-reward rollout. This matters as sampling increases during training and inference, since its benefit depends on retaining probability mass on high-reward outcomes. We propose to optimize this coverage directly. Rather than considering only expected reward, we consider all of its upper tails: for each reward threshold, how likely is the policy to exceed it? This turns a continuous reward into a family of binary success events. We introduce Tail-Likelihood Reinforcement Learning (TailRL), which maximizes the log-probability of exceeding a randomly chosen reward threshold. Its gradient gives more weight to rare, high-reward rollouts and can be interpreted as a mixture of Best-of-(k) gradients. TailRL requires only a simple modification to the advantage function, making it compatible with existing reinforcement learning pipelines. Across object localization, maze navigation, GUI grounding, and code optimization, TailRL leverages rare high-reward training samples to avoid suboptimal solutions and yields models that benefit more from additional samples at inference time.
Training open-ended agents via reinforcement learning (RL) is hindered by the lack of verifiable gold answers and scalable rubrics. Moreover, even near the model's capability boundary, long-horizon open-ended agentic tasks often yield brittle and unstable rewards, resulting in weak or noisy rollout contrast that obscures fine-grained optimization signals for group-based policy learning. To address these challenges, we propose ARISE-RL, a novel full-cycle self-evolution framework that couples a task/rubric Generator and a reasoning Solver through rubric-mediated co-evolution. The Generator grounds tool-related rubric criteria in real tool observations and is rewarded for producing valid, intermediate-difficulty tasks aligned with the Solver's evolving capability boundary. The Solver, in turn, learns from fine-grained rubric satisfaction signals through multi-step reasoning and tool use. We further introduce Reward-Gated Self-Evolution Distillation (RG-SED), which selectively distills a memory-augmented variant of the same policy back into itself only when the memory yields empirical reward improvement, thereby reducing distribution mismatch and avoiding blind imitation of noisy guidance. Finally, to support rigorous evaluation, we present ECR-Bench, an expert-calibrated rubric benchmark suite covering single-tool deep research and multi-tool travel planning. Extensive experiments demonstrate that ARISE-RL consistently achieves robust and stable overall state-of-the-art performance across all evaluated benchmarks.
Rubric-based reinforcement learning decomposes open-ended instructions into prompt-specific, flexible rubrics, making it better suited than reinforcement learning with verifiable rewards for post-training LLMs on open-ended tasks. However, static rubrics are inevitably hacked as the policy evolves, and existing dynamic approaches introduce new problems: undirected rubric extraction, unreliable hack detection, and unbounded rubric proliferation. We propose $\textbf{CARE}$ ($\textbf{C}$ontrastive $\textbf{A}$nchor-based $\textbf{R}$ubric $\textbf{E}$volution), which grounds every rubric evolution step in a high-quality anchor response generated by a frontier model conditioned on the prompt and its rubrics. At each training step, CARE contrasts the highest-scoring rollout against the anchor, enabling two complementary mechanisms: an Adaptive branch that reactively repairs reward misspecification; and a Chase branch that proactively converts frontier-level quality gaps into sharper rubrics. Together, the two branches $\textbf{maintain discriminative accuracy in the high-reward region}$---the precise region where reward over-optimization mostly originates. Experiments on WildChecklist-9K with Qwen2.5-7B-Base and Qwen2.5-7B-Instruct show that CARE achieves state-of-the-art performance on Arena-Hard-2.0, InfoBench, and FollowBench, and is the $\textbf{only}$ method whose win rate against GPT-4.1 anchor responses shows sustained improvement throughout 300 training steps; additional results on Llama-3.1-8B-Instruct and Qwen3-8B further indicate that CARE generalizes across model families.
Codec-based text-to-speech (TTS) models make language-model post-training applicable to speech generation, but it remains unclear when learned perceptual predictors can serve as reinforcement learning rewards without losing alignment with human listeners. We study this question with Group Relative Policy Optimization (GRPO) using learned rewards for anime-like speaking style, naturalness, likability, and arousal. To prevent perceptual rewards from being optimized through transcript drift, we introduce a character error rate (CER) zone constraint and compare policy optimization with Best-of-$N$ reranking under the same reward gate. Across single-reward runs, each reward primarily improves its own target metric, showing that subjective predictors are not interchangeable quality surrogates. Multi-rater A/B tests further show uneven human transfer, while a reward-gap analysis separates average transfer from within-axis calibration: signed reward gaps significantly predict listener choices in the pooled analysis, whereas residual CER gaps do not, but per-axis calibration remains heterogeneous. Best-of-8 is a strong human-level baseline and is not clearly worse than GRPO perceptually, suggesting that GRPO should be viewed as amortizing reward-selected behavior into the policy rather than uniformly outperforming reranking. These results support analyzing subjective speech rewards as predictor-axis-base tuples and provide practical diagnostics for selecting rewards before multi-reward speech post-training.
Current video-to-music (V2M) models lack semantic control and fail to penalize instruction violations, largely due to their reliance on reconstruction objectives and the representational bottleneck of static cross-modal conditioning in Diffusion Autoregressive (DAR) architectures. To resolve this, we introduce VIBE, a novel text-and-video-to-music (T+V2M) generation model that leverages: (1) Conditioning Connection, a depth-wise cross-layer conditioning mechanism that dynamically bridges the planning and diffusion refinement heads and (2) a comprehensive reward modeling taxonomy, optimizing for both hard, verifiable constraints (e.g., tempo, key) and soft, subjective qualities (e.g., musicality, multimodal alignment) with a structured 5-stage training curriculum. Upon evaluation using audio-visual alignment, instruction following, and audio quality metrics, along with a subjective human evaluation study, we observe that VIBE demonstrates enhanced controllability and instruction adherence while performing comparably to most evaluated baselines on generation fidelity and multimodal alignment.
Rubric-based reinforcement learning extends RL beyond tasks with exact answers or rule-based verifiers by scoring responses against instance-specific criteria. However, this makes reward computation expensive: training requires repeated rubric judging, often with proprietary APIs or local generative LLM judges with 7B parameters or more. We study whether smaller language models can serve as efficient and reliable rubric-based judges. To make this question measurable, we construct PointRubric and RaR-Science-Static, two pointwise rubric-based evaluation datasets with instance-specific criteria and itemwise satisfaction labels. We compare three ways of extracting criterion-level judgments from small models: Generative verdicts, Yes/No Logprob margins, and Probe judges. Across both datasets, the Qwen3-1.7B Probe judge achieves the strongest criterion-level agreement among these methods, outperforming Generative and Logprob judges. Used as a GRPO reward model, it trains a policy from 0.232 to 0.643 on RaR-Science rubric score, compared with 0.594 for an 8B Generative judge baseline, while the baseline requires 10.7$\times$ more reward-judge time. Task and domain transfer experiments further suggest that Probe judges preserve criterion-level reward structure across settings.
Reward models play an essential role in aligning visual generative models, yet most existing visual reward models use a single scalar score or rely on fixed criteria that cannot adapt to different instructions. This limits both interpretability and task sensitivity, especially for text-to-image generation and instruction-based image editing, where different inputs require different evaluation dimensions. We propose RubricRM, a pairwise generative reward modeling framework that first produces an input-specific rubric with evaluation dimensions, weights, and scoring criteria, and then applies the rubric to score candidate images. We train dedicated RubricRM models for text-to-image generation and image editing using a two-stage training pipeline: supervised fine-tuning teaches the model the rubric-based scoring paradigm, while GRPO further improves scoring through fine-grained dimension-level rewards. Experiments on multiple generation and editing benchmarks show that RubricRM outperforms existing specialized reward models and remains competitive with strong proprietary MLLM judges despite using smaller backbones. Our models, data, and code are available at https://github.com/zijiankan/RubricRM.
Vision-language models can produce fluent answers that are insufficiently grounded in the visual evidence: a single unsupported object, chart value, or intermediate inference can undermine an otherwise plausible response. We argue that this is a credit-assignment failure in multimodal post-training. Scalar outcome rewards indicate whether an answer is acceptable, but do not identify which visual facts are grounded, which reasoning steps are valid, or which instruction constraints are missed. We introduce Visual Rubrics-Based Reinforcement Learning, which decomposes reference responses into atomic propositions and scores generated answers along Visual Faithfulness (VF), Reasoning Consistency (RC), and Instruction Following (IF). The resulting rubric items provide structured partial credit and localize rubric credit when supporting evidence spans are available. We first obtain an SFT checkpoint by fine-tuning Qwen3-VL-8B-Instruct on the public OpenMMReasoner-SFT-874K corpus, adapting OpenMMReasoner's cold-start data recipe. We construct V-Rubrics 50K, a 50,248-example training set from 17 visually grounded sources, by applying rule-based filters before deriving example difficulty from rejection-sampling scores and then annotating every example with Gemini-3-Pro under the same structured prompt and protocol. We train our model based on the same SFT checkpoint using component-wise, prefix-localized rubric credit. Experiments show that our rubricbased GRPO improves over both the shared SFT baseline and answer-only GRPO, with the largest gains on knowledge-oriented and visually grounded reasoning benchmarks. The results show rubrics as a useful reward abstraction for visual post-training.
Recent studies on GUI agents have increasingly focused on outcome reward modeling, which assigns outcome rewards by judging whether an executed trajectory satisfies the success criteria implied by the user instruction. Existing GUI reward verifiers, however, often under-specify how these criteria should be constructed for each task instance. Whether using generic rubric structures or implicit model reasoning, their judging criteria are not sufficiently task-adaptive: they can transfer checks across tasks, overlook concrete constraints in the current instruction, or become overly strict by enforcing unstated requirements. To address this limitation, we propose AdaptRubric, a Coarse-to-Fine Rubrics Framework that constructs task-adaptive judging criteria through a category-level coarse stage and an instance-level fine stage. AdaptRubric performs category-level coarse rubric retrieval by routing the instruction to a GUI task family and retrieving reusable task-family criteria, then conducts instance-level fine rubric generation to surface compact cues for concrete values, scopes, and constraints in the current instruction. Across offline reward evaluation and online reinforcement learning optimization, AdaptRubric consistently outperforms prior reward agents, improving F1 by 3.6 points over the baseline average under a matched image budget and yielding a 4.23-point task-success gain.
Designing effective reward signals for open-domain question answering is challenging because high-quality responses must simultaneously satisfy multiple aspects of answer quality that are difficult to capture with a holistic scalar objective. We introduce a rubric-based reward framework that generates query-specific rubrics grounded in retrieved evidence and decomposed into multiple quality dimensions, providing fine-grained supervision during post-training. Averaged across three evaluation axes (composition, grounding, and instruction-following), our approach improves over the instruction-tuned baseline by 6.5% and over flat rubric variants by 4%, with consistent gains across all evaluation datasets. Conditioning rubrics on retrieved evidence improves factual support, while decomposing rubrics into quality-specific dimensions further improves coherence, organization, and adherence to query requirements. Our results show that grounded, multi-dimensional rubrics provide more effective reward supervision for complex open-domain question answering.
Reliable reward models are essential for text-to-video evaluation and alignment. However, the trade-off between evaluation accuracy and inference efficiency places high demands on the quality of training supervision. Existing approaches often rely on holistic judges with fixed rubrics or open-ended reasoning, leading to incomplete inspection, unfaithful justification, and entangled attribution. We introduce FIRM-Video, a unified checklist-driven data construction framework based on a check-before-score principle: construct dimension-specific checklists, verify each criterion against temporal visual evidence, and aggregate only verified decisions. For Instruction Following, FIRM-Video decomposes prompts into weighted atomic requirements; for World Coherence, it constructs prompt-calibrated, target-specific checks grounded in visible entities and actions; and for Perceptual Quality, it applies a generic taxonomy of visual defects. The verified criteria and scores are further transformed into natural-language analyses for end-to-end reward modeling. Subsequently, we construct FIRM-Video-90K with 88,044 dimension-specific instances from 29,348 videos, and introduce FIRM-Video-Bench with 750 point-wise human annotations across 250 videos. The Qwen3-VL-based FIRM-Video-8B achieves the best overall MAE on FIRM-Video-Bench while consistently delivering the highest VBench Total, Quality, and Semantic Scores in Best-of-8 sampling across three video generators.
Bogdan Zagribelnyy, Ivan Ilin, Nikita Bondarev +5cs.LG cs.AI cs.CE cs.CL
Single-step retrosynthesis is a central component of computer-aided synthesis planning, yet its intrinsically one-to-many nature is poorly captured by single-answer evaluation and benchmarking protocols. To address this, we introduce Top-K prompting as a robust training and inference paradigm to better capture diverse, plausible reaction predictions. We compile CREED-CCV-2+USPTO-XL, an ultra-large-scale dataset of ~45.6 million verified reactions to train the C3LM (Chemistry Constraint-Consistent Language Model). By integrating fine-tuning with ChemCensor-based and novelty-oriented rewards, our model achieves state-of-the-art performance on the OOD URSA-expert-2026 benchmark. Further analysis of reaction uniqueness shows that LLMs and conventional models explore complementary reaction spaces, motivating ensemble-based retrosynthesis systems. Overall, our results establish Top-K, plausibility-aware training as a practical new direction for robust future LLM-based synthesis planning.
Using reinforcement learning to post-train joint video-audio generation models requires a reward signal. Existing methods construct this reward by combining metrics for individual quality dimensions, including audio quality, visual fidelity, and synchronization. However, these metrics evaluate perceptual dimensions separately and fail to capture the overall semantic and temporal coherence among the text prompt, video, and audio that shapes human preferences. Optimizing models against these metrics encourages reward hacking, generating video-audio content that achieves high scores on these metrics yet appears incoherent or unfaithful to human viewers. To address this problem, we first construct a large-scale human-preference dataset VAPref-10K for joint video-audio generation, comprising 9K prompts and 10.3K fine-grained paired comparisons from open-source generation models. We also introduce the VA-Judger-Bench benchmark with both in-domain and out-of-domain model comparisons to evaluate whether reward models truly align with human preferences. We further propose VA-Judger, a chain-of-thought omni-reward model for joint video-audio generation. In particular, VA-Judger first learns from pairs with clear quality gaps to establish structured output and coarse preference discrimination, then distills reliable preference explanations for harder near-quality comparisons via rejection sampling verified against human annotations, and finally performs dimension-wise reinforcement learning that decomposes human feedback into individual quality dimensions for denser reward signals than a single binary preference label. Experiments show that VA-Judger outperforms metric baselines in predicting human preferences on both in-domain and out-of-domain evaluations. Using its human-aligned rewards for post-training audio-video generation model also yields significant improvements in generation quality.
Vision-language-action (VLA) models improve robotic manipulation but remain vulnerable to compounding errors, scene changes, and off-trajectory states. Reinforcement learning can refine pretrained VLA policies, yet sparse success signals hinder exploration, while engineered dense rewards are costly and task-specific. Existing learned visual reward models often rely on static before-after observations, causing temporal ambiguity and weak discrimination between robustness-preserving variations and task-invalid failures under out-of-distribution (OOD) execution. We introduce Robo-Dopamine 2.0, a history- and OOD-aware process reward model with a pairwise prediction interface. It combines (1) history-conditioned pairwise rewards that use source-aligned reference panels for synthetic OOD queries and observed rollout history for online queries, while preserving the queried endpoints, and (2) an OOD-aware signed progress space that represents valid progress, robustness, failure, and recovery. A Signed-Hop Curriculum with transition-aware replay learns coarse execution ordering before fine-grained progress calibration. We also construct an OOD trajectory dataset and a five-family benchmark. Reference panels improve mean visual order consistency (VOC) from 0.967 to 0.986 and OOD-robust VOC from 0.906 to 0.958. With the same 400K pairwise-reward budget, Signed-Hop training with 25% replay reaches 0.9872 mean VOC, compared with 0.9858 for a matched-pool shuffled control. In downstream reinforcement learning, the full model achieves 86.8% mean RoboTwin success and 71/80 successful real-world insertions.
Video generation is central to AI-powered content creation. Aligning generated videos with human preferences is a key criterion for evaluating generation quality. Despite significant progress in visual quality, three key challenges remain. First, the reliability of reward signals is constrained by the quality of human preference data, which is often affected by subjective noise and bias. Second, standard scalar reward models collapse multi-aspect human preferences into a single value, leading to the loss of dynamic trade-offs across multiple preference dimensions. Third, in policy optimization, the widely adopted KL divergence imposes primarily local constraints and may fail to capture the global structure of human preferences. To address these challenges, we propose a unified preference-aware learning framework for video generation. First, we introduce elite-guided filtering to calibrate preference data and construct reliable supervision for reward model training. We then model video quality as a multidimensional reward distribution to capture the uncertainty inherent in human preferences, and use the Wasserstein distance to align the learned reward distribution with the empirical human preference distribution. Finally, we introduce Wasserstein-based distributional alignment into GRPO, guiding policy optimization to better match the global structure of human preferences over videos. Experiments on reward modeling and video generation demonstrate that our approach improves the reliability of reward signals and the perceptual consistency of generated videos. Our code is available at https://github.com/alignhs26/ahs.
Generative model alignment has received broad interest, and significant progress has been made in supervised fine-tuning and inference-time computation. Yet, alignment has remained poorly understood from a statistical learning perspective. We formulate inference-time alignment as a weak-to-strong learning problem, where a reference policy (weak model) is assumed to be fairly good and the goal is to produce a strong model that predicts a good response at test time with arbitrarily high probability. Our problem is formulated as learning from scratch --- everything is learned from data rather than assuming access to a good reward estimate, and thus differs from the existing inference-time alignment theory. Our framework shares similarity to the recent work of Joshi et al., (arXiv:2510.15464), where for each prompt, there could be multiple good responses. Our definition of the alignment learnability follows the standard PAC learning principle. We introduce a novel combinatorial dimension of the reward class which we call the alignment dimension, and show that it completely characterizes the alignment learnability --- a reward class is alignment learnable if and only if its alignment dimension is finite. The core of our learning procedure works by learning a pairwise comparator and then running a tournament over candidate responses. We believe that our results might shed light toward establishing a complete theoretical understanding of alignment.
Reinforcement learning (RL) for terminal agents needs executable training environments with reliable rewards and useful difficulty. Fixed recipes such as few-shot, Self-Instruct, and Evol-Instruct apply the same prompting policy to every seed, even when the current policy would benefit from a harder, easier, or simply different task. We present Envs-FORGE, a prompting policy that converts verifier rewards into per-seed environment-synthesis actions. Envs-FORGE estimates seed pass rates, scores six projection--direction actions around a target learning frontier, and solves a per-seed mixed-integer linear program (MILP) to choose the action that conditions generation. The selected action drives synchronized rewriting of the instruction, fixtures, oracle solution, tests, and Docker environment; only gold-verified bundles enter RL training. The indexed MILP form also supports optional soft skill coverage for portfolio planning. On Qwen 3.5 35B, Envs-FORGE improves Pass@1 over Base by 9.2 percentage points on tb-core (40.0% to 49.2%) and 6.4 points on tb-2.0 (23.0% to 29.4%), exceeding the strongest fixed-recipe baseline by 2.4 and 2.1 points. It reaches 77.1% on SWE-bench Verified versus 73.4% for Base, and improves tb-core by 6.8--9.2 points across the evaluated 4B--35B models. All synthesis methods export 100 verified environments and use 2.27M--2.88M synthesis tokens, placing the comparison at the same downstream training-set size and the same operational scale. The source code is available at https://github.com/DataArcTech/DataArc-SynData-Toolkit/.
Yongqi Tong, Tan Li Hui Faith, Choy Zhen Wen Marcus +5cs.AI cs.CL
Open-ended real-world interaction admits multiple valid behaviors: an agent may answer directly, ask for clarification, provide progress updates, or confirm before acting. This flexibility breaks a core assumption behind group-based RL: rollouts compared within a group are no longer guaranteed to be behaviorally comparable. As a result, reward-model preferences over interaction style can distort relative advantages and steer optimization toward reward-preferred behaviors rather than context-appropriate ones. We formalize this as a \textit{reward fairness problem} and propose \textbf{ARC} (Advantage Regularization via Conditioning), a training recipe that restores fairer relative comparison through strategy-conditioned rollout grouping, together with hybrid rewards and entropy regularization. We study ARC in our proposed \inter, a novel paradigm for responsive, steerable, and execution-aware user-agent interaction that decouples user-visible communication from latent reasoning and tool use. \inter\ also provides the annotation and distillation pipeline for constructing \inter-86K, our strategy-annotated training corpus for supervised and RL training. Empirically, ARC substantially strengthens the core $τ/τ^2$ tool-use benchmarks, while \inter\ reduces time-to-first-token from 4.91s to 1.27s relative to a think-style baseline. Together, these results suggest that a central bottleneck in open-ended interactive learning is not only how agents are rewarded, but whether their behaviors are compared fairly in the first place. The ARC implementation and \inter-86K training data will be released.
Jiabao Zhuang, Changhao Jiang, Hanchen Wang +11cs.SD cs.CL
Long-form song generation models continue to improve in duration, structural integrity, and acoustic complexity, making reliable aesthetic rewards increasingly important for aligning these models with human preferences. However, reward models for complete songs remain limited, and existing evaluators typically predict scores in a single forward pass without providing readable explanations. We introduce MUSECRITIC, a semi-scalar reward model that generates a natural-language critique covering five aesthetic dimensions and uses it as an intermediate representation to predict continuous reward scores. MUSECRITIC follows a two-stage training pipeline: a teacher model first provides high-quality critiques for supervised fine-tuning, after which the fine-tuned model generates its own critiques for reward learning, mitigating distribution shift between training and inference. On an in-domain test set of 200 SongEval songs, MUSECRITIC reduces macro-averaged mean squared error from 0.2875 to 0.2316 and improves macro-averaged LCC, SRCC, and Kendall's tau to 0.9068, 0.8838, and 0.7178, respectively. On the out-of-domain Music Arena benchmark with 733 preference pairs, it achieves the highest accuracy of 71.35%. Moreover, using MUSECRITIC with GRPO improves Muse-0.6B on all nine aesthetic metrics from SongEval and Audiobox Aesthetics. These results demonstrate that critique-conditioned reward modeling reduces scoring error and provides an effective optimization signal for song generation. The project repository is available at https://github.com/WuqnEl/MuseCritic.
M P V S Gopinadh, Karthik Kamuju, Kummari Avinash +2cs.LG cs.AI cs.CL
Reinforcement Learning from Human Feedback (RLHF) aggregates heterogeneous preferences into a single reward model, assuming preference homogeneity. When preferences are heterogeneous, this aggregation induces a procedural fairness failure where majority preference groups dominate reward learning while minority preferences are systematically under-represented. This work defines procedural fairness in alignment as preserving distinct preference signals during reward modeling and shows that standard RLHF violates this via preference averaging. Preference-Aware RLHF (PA-RLHF) is introduced, separating optimization across preference modes at the reward learning stage. In a controlled setting, PA-RLHF improves overall alignment accuracy from 46.9% to 67.9% and reduces the fairness gap between best and worst aligned groups from 15.9 to 9.6 percentage points. These results show that procedural fairness failures in alignment can arise from structural design choices in reward learning, even in controlled, noise-free settings, with direct implications for large language models and agentic systems, where biased reward models can compound inequities across sequential decisions.
Multi-agent reinforcement learning (MARL) is a powerful framework for solving complex collaborative tasks, but it relies heavily on well-defined global reward functions. Designing such rewards is challenging, especially in systems with heterogeneous agents, where a single scalar objective may fail to capture diverse behaviors. In this paper, we introduce Multi-AGent Preference-Integrated lEarning (MAGPIE), which addresses these challenges through agent-specific preference modeling. Each agent is evaluated by a dedicated expert through preference signals, eliminating the need for global evaluation. We theoretically prove that optimizing these decentralized preferences converges to a Nash equilibrium policy. To integrate local preferences into a coherent global objective, we construct agent-specific reward models from preference data and combine them via a monotonic aggregation mechanism. We further prove that optimizing this aggregate reward model is equivalent to training the Nash equilibrium policy. Extensive experiments on benchmark multi-agent tasks and a sequential production line task show that MAGPIE achieves performance comparable to reward-engineered baselines, demonstrating its potential to facilitate policy learning in scenarios where precise reward engineering is impractical.
Reward models are a bottleneck for reinforcement learning in embodied AI. Long-horizon robotic manipulation requires scalable vision feedback beyond handcrafted rewards or task-specific annotations. Existing open-source VLM reward judges like RoboReward adopt simple 1--5 trajectory progress scoring, lacking pairwise preferences for RLHF, DPO and Bradley-Terry frameworks, while failing to optimize video scene understanding. Augmenting RoboReward with pairwise comparison and video-QA supervision causes inconsistency between pairwise preferences and pointwise scores, introducing training noise and hurting downstream performance---an issue aggregation methods such as TrustJudge cannot resolve. To address this, we propose TrustRoboReward, a multi-paradigm reward modeling framework equipped with Preference-Ordered Isotonic Score Editing (POISE). We construct a unified four-paradigm dataset with trajectory progress scoring (Score-A), video-QA answer quality scoring (Score-B), and their pairwise counterparts (Pair-A, Pair-B). Pairwise labels align better with human judgment than pointwise scores, inspiring us to calibrate pointwise scores to avoid score-pair reversals against pairwise preferences. POISE rectifies pointwise scores and eliminates cross-paradigm reversal conflicts unresolved by TrustJudge. Theoretically, POISE reduces score-pair reversal conflicts from 20.15% to 0%, whereas TrustJudge retains 20.46% conflicts on the same corpus. Evaluated on our benchmark, Qwen3-VL-4B trained with POISE achieves an overall reward score of 77.96%, nearly matching GPT-5-mini (78.09%, gap 0.13%) and outperforming the strongest RoboReward-4B baseline by 10.13%. It also lifts test-time score-pair consistency to 71.90%, exceeding RoboReward-4B (57.26%) and GPT-5-mini (68.09%). Integrating TrustJudge aggregation during inference boosts the overall score to 78.57%, surpassing the GPT-5-mini teacher model.
Chenglong Wang, Ziming Zhu, Yifu Huo +9cs.LG cs.CL
Recent advances in reward modeling show a paradigm shift from discriminative reward models to generative reward models. However, despite their strong capabilities in response ranking, generative reward models have not realized their potential in reinforcement learning (RL). Our analysis reveals that this limitation arises from a mismatch between the comparative nature of generative reward modeling and the scalar scoring paradigm adopted by existing RL algorithms. To bridge this gap, we propose a Ranking-based Reward Construction (RRC) approach, which enables generative reward models to provide more effective RL learning signals by deriving rewards from relative preference rankings. RRC introduces two complementary strategies: self-competitive ranking, which exploits comparisons among sampled responses, and anchor-guided ranking, which enables scalable ranking-based reward construction with a small set of reference responses. Experiments across open-ended chat and reasoning benchmarks demonstrate that RRC substantially improves RL training with generative reward models, achieving consistent gains over existing reward construction approaches. Our code can be found at https://github.com/wangclnlp/RRC.
Aligning diffusion models with human preferences usually relies on a sparse terminal reward evaluated on the final generated samples, presenting a severe temporal credit-assignment challenge across the multi-step denoising process. We propose Latent Reward Registers, a mechanism that estimates terminal preference directly from intermediate noisy latents by prepending learnable, position-free register tokens to the input sequence of a frozen Diffusion Transformer (DiT). This independent readout mechanism extracts latent reward evidence without altering the generator's hidden states or velocity field. The resulting dense, differentiable reward signal throughout the full denoising process facilitates two alignment strategies. For training, Reward-Gradient On-Policy Distillation (RG-OPD) distills reward-guided updates along on-policy trajectories, bypassing the computationally expensive rollouts of standard policy gradients. For inference, Reward-Guided Sampling (RGS) steers trajectories via magnitude-matched reward gradients without parameter updates. Empirically, at high noise levels (u = 0.8), the registers reach the highest pairwise accuracy among the evaluated latent reward models. Furthermore, RG-OPD outperforms online reinforcement learning baselines while reducing GPU hours by up to 33x, and RGS establishes a new state-of-the-art among training-free methods, strictly enhancing both alignment and perceptual metrics. Code and weights are available at https://github.com/Guanys-dar/latent-reward-register
Seongyoon Kim, Boryeong Cho, Jihwan Oh +2cs.LG cs.AI
Large language models are increasingly aligned to human preferences via reward modeling, but user preference data are sensitive and often cannot be centralized. Federated learning keeps such data local while learning a shared initial reward model, which is later personalized for each client through local fine-tuning. Because users often assign opposite labels to the same pair of responses, existing federated methods address preference heterogeneity by clustering similar clients and training one reward model per group, assuming that each group requires its own initialization. We show that this assumption is unnecessary. Under balanced preference groups, a single FedAvg model, despite starting at nearly random accuracy, surpasses reward models trained separately for each ground-truth group after only a few local optimization steps. We attribute this phenomenon to the flatness of the shared initialization: averaging across all clients learns richer shared representations that distinguish responses while canceling conflicting preference directions, leaving the model near a decision boundary that can be rapidly adapted. Group imbalance breaks this effect as the cancellation becomes asymmetric and leaves minority clients too far from the boundary to recover. Motivated by this observation, we propose FedGD (Federated Learning with Group Debiasing), which discovers latent preference groups during federated training and learns a single reward model using group-debiased client sampling. By counteracting the effect of group imbalance, FedGD learns an initialization that remains highly adaptable, enabling effective personalization without prior knowledge of the underlying groups.
Large language models are being extensively used to simulate individual user behavior, yet faithfully representing a population requires capturing the systematic variation in values, beliefs, and cultural norms that distinguish one group from another. We introduce Population Aligned Language Models (PALMs), a suite of models each aligned to specific populations, covering five countries: USA, India, Brazil, France and Italy. PALMs are created by synthesizing rationales grounded in psychological and cultural constructs and using these as latent supervision during preference tuning for population-specific alignment. Evaluated across four dimensions: personality, values and beliefs, cultural norms, and morality, PALMs consistently outperform baselines, including culture-specialized models, achieving an average of 8.59% relative improvement over the best baseline across all five populations. Notably, construct-grounded rationales outperform both demographic prompting and survey-based fine-tuning, suggesting that grounding preference learning in psychology and culture provides a richer inductive signal than surface-level response distributions. We further demonstrate strong generalization to downstream applications with- out task-specific supervision: outperforming best baselines by 5.19% in personalized reward modeling, 6.34% in population simulation, and showing strong transfer to social reasoning tasks. Datasets and code are available at: https://github.com/limenlp/PALMs.