Reasoning-oriented large language models often achieve strong problem-solving performance by generating long chains of thought, but this behavior substantially increases inference cost and latency. In contrast, instruction-tuned models tend to answer more concisely, yet often lack comparable reasoning ability. This accuracy-efficiency mismatch motivates a lightweight approach that combines the strengths of both models without full model retraining. In this paper, we propose GRIP (Granular Reward-guided Interpolation of Parameters), a reward-guided parameter interpolation framework for efficient reasoning. Given a reasoning model and an instruction model with identical architectures, GRIP assigns learnable interpolation ratios to individual modules and optimizes only these ratios while keeping both source models frozen. The interpolation ratios are trained with a reward signal that favors responses that are both correct and concise. Experiments show that GRIP achieves a better accuracy-efficiency trade-off than fixed or search-based merging baselines and further reveals module-wise fusion patterns associated with efficient reasoning.
Xiaoying Song, Anirban Saha Anik, Jinyu Liu +3cs.AI
Correcting health misinformation in dialogue requires more than producing a factual rebuttal: users differ in what they know, what they believe, and what they need to hear, so an effective intervention often depends on first asking the right clarifying question. Yet existing methods either respond immediately or probe indiscriminately, treating clarification as either unnecessary or always beneficial. We propose Reward-Optimized Probe-and-Respond (RO-PnR), a framework that learns when asking is worth its cost. At each turn, RO-PnR chooses between probing for more information and committing to a final correction, guided by a turn-level reward that weighs the expected gain from probing against its interaction cost. To capture how user heterogeneity affects probing value, we model each simulated user with a latent state along health literacy and belief commitment. Experiments show that RO-PnR achieves the highest cost-adjusted utility across three health-misinformation datasets and three base models, using 30% fewer turns than always-probe baselines.
Text editing in product posters entails inserting new text or replacing existing text while preserving product appearance, background content, and global composition. Despite recent progress in instruction-based image editing, general-purpose models remain unreliable in this setting: they often omit or incorrectly render the target text, place it over salient products or pre-existing content, and produce structurally distorted or visually inconsistent glyphs. We introduce \textbf{TextRefine}, a task-aligned post-training framework that combines supervised fine-tuning with operation-specific reward optimization to address these complementary failure modes. For text insertion, our text-span-level reward jointly assesses semantic fidelity and target-span coverage, penalizes spatial conflicts with products and existing text, and employs a gated structural constraint to preserve non-text regions. For text replacement, our glyph-level reward leverages the connectionist temporal classification (CTC) posterior of the target character to provide graded supervision for fine-grained defects, including missing strokes, structural deformations, and confusion among visually similar characters. We further introduce \textbf{OpenTextEdit}, a dataset comprising 100K images for text editing in product posters, with multi-text layouts, detailed text attributes, product masks, and challenging low-frequency characters. Extensive experiments on both insertion and replacement demonstrate that TextRefine consistently outperforms the evaluated image editing baselines in textual fidelity, placement reliability, and glyph quality while better preserving source-image content.
Learning a reward model from human feedback and optimizing a policy against it is one approach to aligning AI systems with individual users. From a fairness perspective, existing work improves such alignment by developing data-efficient and accurate reward models that capture minority preferences despite scarce data. We push this line of inquiry one step further and argue that data-efficient and accurate per-user reward models are not sufficient: users whose reward models are difficult to \textit{optimize} at the policy level can become a new underserved group. We start from the observation that one user's reward model can be easy to optimize from the initial policy while another's is not. We argue that, given a sufficiently diverse user population, a curriculum naturally emerges between easy- and hard-to-optimize reward models. Building on this insight, we propose CurriPO, which grows a tree-structured curriculum to accommodate diverse user-specific objectives, covering the population in a single traversal. Specifically, CurriPO automatically constructs a curriculum over diverse user reward models, allowing it to branch from the existing curriculum and reuse reward models previously incorporated into the curriculum. To the best of our knowledge, this is the first work to explicitly exploit multi-user structure to address optimization in AI alignment. Extensive experiments on personalized continuous control in a simulated environment show that CurriPO achieves $1.2$--$2.1\times$ the population satisfaction of the strongest baseline while substantially reducing training time. Additional analysis attributes much of this improvement to the users left underserved by conventional optimization.
Hojun Choi, Jaeyo Shin, Suin Lee +1cs.AI cs.CV cs.LG
Agentic systems powered by large language models (LLMs) have opened new opportunities for business ideation. Yet existing approaches remain confined to a text-only paradigm, despite the inherently multimodal nature of real-world contexts. We thus introduce MBA-Bench, the first multimodal benchmark for training and evaluating business ideation agents, comprising 30K samples across six domains, each domain characterized by distinct visual cues not fully conveyed by text alone. Concretely, we automatically caption images and employ GPT-4o to generate five reference ideas for each of three business questions through retrieval query generation, market evidence retrieval, and evidence-augmented synthesis. Following prior work, we evaluate agents across six business-oriented criteria using MLLM-as-a-Judge. To consider settings where criteria are hidden or disclosed, we present MBA-b and MBA-k for blind and known, respectively. We train both with two novel reward objectives---creativity and feasibility---while MBA-k further optimizes the six disclosed criteria for eight in total. Both are trained via LoRA-based supervised fine-tuning followed by group relative policy optimization with these setting-specific rewards. For extensive experiments on MBA-Bench, we set up two baselines accommodating either captions only or multimodal inputs, with the latter nearing closed-source performance on several metrics. MBA-b and MBA-k outperform caption baselines by 63.9% and 77.1%, and multimodal baselines by 25.6% and 35.8%, respectively.
Diffusion models have strong generative capabilities. However, their maximum likelihood training objective only focuses on reconstructing the data distribution, making it difficult to align with specific preferences. Reinforcement learning (RL) for preference alignment in diffusion models is promising but limited by reward sparsity. Since a single reward cannot support optimization, existing RL methods usually backpropagate the final reward to all previous steps. However, denoising is stage-wise, with distinct semantics and controllability. Repeating the final reward across all steps creates a temporal objective mismatch, encouraging reward shortcuts that lead to reward hacking. At the same time, due to reward backfilling, each time step receives the same reward, making it impossible to distinguish between actions, thereby weakening the optimization process. To resolve this issue, we propose Stage-Guided Per-Step Optimization (SGPO) for diffusion models, which jointly leverages signal-to-noise ratio and semantic changes to identify generation stages and adaptively assign stage-specific objectives. Early denoising is chaotic and far from the final reward, resulting in weak reward-behavior correlation. This stage should prioritize exiting the chaotic state. In the mid stage, the latent transitions to a stable structure, where the final reward better corresponds to generative behavior. Therefore, this stage optimizes the final reward while exploring diversity to avoid early convergence to a single mode. In the late stage, the latent's core structure is largely fixed, and preference optimization mainly amplifies local details, risking overfitting. Therefore, stable convergence is preferred to avoid quality degradation. Results from 16 comparative experiments validate SGPO. Our method achieves 26.7% average gains in generative quality and 36.7% higher convergence speed.
Yingqing Guo, Hui Yuan, Zijian He +2cs.LG cs.AI cs.CV
Flow-based generative models are typically sampled by solving a deterministic ordinary differential equation (ODE), whereas online reinforcement learning requires stochastic rollouts for policy exploration and optimization. Existing GRPO methods for flow models therefore replace the inference-time ODE with a stochastic differential equation (SDE) during training. Although the ODE and SDE share the same marginal distributions in continuous time, their finite-step discretizations can differ substantially. In particular, SDE rollouts often become blurry as the exploration noise increases, creating a mismatch between the samples used for reinforcement learning and those generated by the test-time ODE sampler. We introduce LC-GRPO, a flow-based GRPO framework with Langevin correction. Each rollout transition first takes an inference-aligned ODE Euler step and then applies a stochastic Langevin correction targeting the marginal distribution at the resulting timestep. The required score is recovered directly from the flow velocity, requiring no additional score model, while the resulting transition remains an isotropic Gaussian with a tractable likelihood for policy optimization. We theoretically show that, under suitable conditions, one Langevin correction step reduces the Wasserstein error of an imperfect ODE Euler step. At a matched randomness level, we further show that the proposed transition can be more accurate than the standard Euler--Maruyama discretization of the reverse SDE. Experiments on SD3.5-Medium, FLUX.1-Dev, and HunyuanVideo demonstrate that LC-GRPO consistently improves reward optimization across text-to-image and text-to-video tasks, preserves generation quality, and substantially narrows the gap between stochastic training rollouts and deterministic test-time ODE inference.
Supervised fine-tuning (SFT) can equip large language models (LLMs) with domain knowledge for high-performance computing (HPC) tasks such as data race detection and benchmark question answering. However, knowledge alone does not guarantee task-appropriate behavior: the same SFT model that correctly classifies 88.65\% of C/C++ data race samples produces verbose, imprecise answers to factual queries, with 65.9\% of MLPerf responses exceeding 40 characters. Reinforcement learning (RL) post-training addresses this gap by optimizing for task-specific rewards rather than token-level imitation. Yet HPC tasks exhibit extreme heterogeneity, with binary classification, factual QA, and semantic generation differing by 58x in answer length, spanning three distinct reward distributions, and showing widely varying SFT accuracy. This makes uniform-weight RL methods such as GRPO suboptimal. We propose HARGO, Heterogeneity-Aware Reward-Guided Optimization, which introduces per-response importance weighting via confidence-modulated advantage: computing a discrimination signal from group-level reward contrast and a confidence signal from reference model log-probabilities, then modulating the advantage before computing per-response weights, without requiring task-type labels. Across four HPC tasks and nine methods, HARGO achieves the best performance on all three primary metrics: WinRate 54.62\%, Data Race F1 91.30\%, and PLP Similarity 0.8558. Ablation confirms complementary contributions from both signals. HARGO establishes the best overall alignment quality among compared methods for heterogeneous HPC tasks.
Large-scale video diffusion models (VDMs) deliver strong generation performance, but full fine-tuning for downstream tasks incurs prohibitive computational costs. Existing parameter-efficient fine-tuning (PEFT) methods have two critical flaws on billion-scale models: they still require substantial trainable parameters, and reward-based training suffers from noise-induced optimization instability in condition-guided tasks. We propose MagicPrompt, a lightweight framework that achieves extreme parameter efficiency and stable reward optimization. It first adopts Attention-Embedded Prompt Tuning, which steers generation via lightweight soft prompts with orders of magnitude fewer parameters while preserving pre-trained knowledge. It further introduces Dual-Space Reward Feedback Optimization, which uses self-supervised latent objectives to improve condition-guided reward training. Experiments show MagicPrompt reaches competitive performance with less than 1\% trainable parameters and notably reduces training costs.
Sihang Nie, Xiaofen Xing, Rui Xing +5eess.AS cs.CL cs.SD
Recently, Large Language Model (LLM)-based Text-to-Speech (TTS) models have achieved remarkable naturalness. However, the standard Supervised Fine-Tuning paradigm often converges to statistically averaged prosody, limiting emotional expressiveness. While preference-driven optimization offers a promising alternative, existing approaches suffer from two structural mismatches: information conflict, where content and emotion in a shared latent space produce conflicting gradients, leading to reward hacking and semantic degradation; and scale gap, where sparse sentence-level rewards struggle to guide dense frame-level generation. To overcome these challenges, we propose HPRO, a hierarchical progressive reward optimization framework. Within HPRO, we introduce the HD-Emo codec as a novel differentiable reward model to resolve the information conflict. It extracts speech into distinct content and style preference tokens, structurally isolating emotional optimization from semantic content. Building upon this structured preference space, HPRO bridges the scale gap by progressively aligning frame-, word- and sentence-level objectives. Experiments demonstrate that HPRO significantly enhances emotional expressiveness, while effectively preserving linguistic intelligibility. The code and audio samples are publicly available at https://xxh333.github.io/hpro-demo/.
Wenwang Huang, Yusen Fu, Junjie Wang +6cs.CV cs.AI
While personalized image generation has achieved remarkable progress, multi-reference image generation (MRIG) remains a challenging task. Most existing benchmarks fail to adequately evaluate complex MRIG scenarios, hindering further progress in this area. To better assess model performance on complex MRIG tasks, we introduce OmniRef-Bench, a benchmark that covers complex combinations of reference image types and a large number of reference images. Evaluations on OmniRef-Bench show that mainstream open-source models struggle in complex MRIG scenarios, and their performance deteriorates significantly as the number of mixed-type reference images increases. To address this issue, we propose DyRef, a two-stage training framework. In the first stage, supervised fine-tuning equips the model with the basic capability to handle complex MRIG tasks. In the second stage, we introduce Difficulty-aware Advantage Reweighting (DAR) and Discriminative Reward Scaling (DRS). DAR dynamically adjusts the optimization objective to improve performance when handling a large number of mixed-type reference images. DRS enlarges intra-group reward differences for more effective policy optimization. Experiments demonstrate that DyRef significantly improves the performance of open-source models on OmniRef-Bench and single-image editing benchmarks, demonstrating the effectiveness and generalization capability of our approach.
Protein language models (PLMs) have emerged as powerful tools for controllable biomolecular design, yet their post-training adaptation typically relies on costly wet-lab validation or curated preference datasets. To overcome this supervision bottleneck, we introduce unsupervised reward optimization of PLMs, a comprehensive framework for steerable protein generation without ground-truth labels. Our key insight is that task-agnostic rewards, which combine intrinsic model uncertainty with extrinsic semantic consistency informed by protein representation models, exhibit strong correlation with controllability measures across base models and temperature regimes. Building upon this discovery, we propose two offline algorithms: Soft Reward Optimization (SRO) and Binarized Reward Optimization (BRO), which effectively maximize the classical RLHF objective induced by these proxy rewards. Extensive experiments on compositional out-of-distribution prompts demonstrate that both methods significantly outperform competitive baselines (DPO, KTO), while approaching oracle performance across multiple sampling temperatures, model scales and protein families. Moreover, PLMs fine-tuned with unsupervised rewards can achieve consistently higher coverage compared to their base model in pass@k evaluations. By enabling self-improvement of PLMs through their own generated experience, our framework provides a scalable pathway toward controllable biomolecular design in settings where labeled preferences or experimental feedback are scarce or unavailable.
We propose GLASS, a framework for composable acoustic style control in zero-shot autoregressive text-to-speech (TTS) that learns controls from post-generation rewards rather than style labels. In zero-shot TTS, a speaker prompt often entangles speaker identity with prosodic attributes such as speaking rate and pitch, making it difficult to change style without changing the prompt itself. GLASS instead treats each acoustic attribute as a reward-defined control direction. For each control axis, GLASS freezes the TTS backbone and trains one lightweight LoRA adapter with Group Relative Policy Optimization (GRPO), using speech-token length and mean F0 as style rewards and WER as an intelligibility anchor. Because each control is represented as a LoRA weight update, independently trained adapters can be swapped, interpolated, and composed through linear LoRA arithmetic without retraining the backbone. Experiments on speaking rate and pitch control show targeted style shifts while preserving naturalness, speaker similarity, and intelligibility, and demonstrate smooth interpolation and multi-axis composition across independently trained adapters.
Modern video diffusion models excel at appearance synthesis but still struggle with physical consistency: objects drift, collisions lack realistic rebound, and material responses seldom match their underlying properties. We present PhyCo, a framework that introduces continuous, interpretable, and physically grounded control into video generation. Our approach integrates three key components: (i) a large-scale dataset of over 100K photorealistic simulation videos where friction, restitution, deformation, and force are systematically varied across diverse scenarios; (ii) physics-supervised fine-tuning of a pretrained diffusion model using a ControlNet conditioned on pixel-aligned physical property maps; and (iii) VLM-guided reward optimization, where a fine-tuned vision-language model evaluates generated videos with targeted physics queries and provides differentiable feedback. This combination enables a generative model to produce physically consistent and controllable outputs through variations in physical attributes-without any simulator or geometry reconstruction at inference. On the Physics-IQ benchmark, PhyCo significantly improves physical realism over strong baselines, and human studies confirm clearer and more faithful control over physical attributes. Our results demonstrate a scalable path toward physically consistent, controllable generative video models that generalize beyond synthetic training environments.