Preference Optimization

Follow this topic Ask about this topic

Momentum

13 papers in the last four weeks, up 30% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 98

Oct 7, 2026cs.AI

Successive Training Stages and Large Language Model Persuasion: Effects of Misalignment, Supervised Fine-Tuning, and Preference Optimization

Large language models (LLMs) can be tuned to influence human attitudes, yet the respective contributions of successive post-training stages remain un-clear. This study examines how three successive training stages affect LLM persuasiveness: (1) misalignment through supervised fine-tuning (SFT) on conspiracy data, (2) additional persuasive SFT on argumentative data, and (3) Identity Preference Optimization (IPO), a preference-optimization method. A total of 835 participants recruited on Prolific were randomly assigned to five between-subject conditions (neutral text, conspiracy-trained model, persuasion-trained model, preference-optimized model, and GPT-4) and were exposed to texts on 10 divisive political issues, personalized from their individual profiles in all model conditions. Attitude change was measured as the difference between pre- and post-exposure positions on continuous Likert scales and analyzed with an analysis of covariance (ANCOVA). A significant condition x baseline-attitude interaction, F (4, 825) = 5.33, p < .001, indicated that training effects depended on participants' initial attitudes. Persuasive SFT produced greater attitude change than conspiracy training alone, d = 0.30, whereas IPO provided no additional benefit, d = 0.03, and GPT-4 did not differ from neutral text, d = --0.01. These results show that targeted supervised training on persuasive data increases LLM persuasiveness, whereas preference optimization yields no significant gains beyond it.

Antony Dalmiere, Pascal Marchand, Guillaume Auriol +1

LAAS-TRUST · LAAS-TRUST, INSA Toulouse · LAAS-TSF, LAAS

Persuasion in Language Models LLM Post-Training Supervised Fine-Tuning Preference Optimization

Oct 7, 2026cs.RO

RobotAPO: Adversarial Physics Preference Optimization for Robotic Manipulation Video Generation

Robotic manipulation videos are increasingly used as visual plans for embodied agents, but optimizing purely for visual plausibility often fails to capture the fragile physical manifold of real-world interactions. Even minor physics-violating errors at the interaction boundary, such as interpenetration or premature object motion, can completely invalidate the inferred timing and pose needed for downstream execution. Because standard supervised fine-tuning lacks the direct pressure to penalize these localized failures, we introduce AgiBot-PhysPref. This rigorously curated 10,000-sample preference dataset isolates condition-matched physics violations, turning the generator's own failure distribution into a foundational signal for physical consistency. Building upon this, we propose RobotAPO, an adversarial physics preference optimization framework operating in the continuous flow-matching denoising space. To prevent the policy from merely memorizing static curated failures, RobotAPO employs a lightweight adversarial counterfactual proposer that learns a condition-dependent, physical-failure-biased direction in denoising space. This encourages the model to explore and better respect the physical interaction boundary, all while maintaining a pure prompt-and-reference inference interface without requiring external structural conditioning. Comprehensive evaluations demonstrate that explicitly correcting these localized physics violations improves downstream robot execution from generated videos. On held-out AgiBot conditions, RobotAPO outperforms the strongest controlled internal baseline in physical consistency by 6.8% hard score and 10.0% soft score. Crucially, in real-robot replay, it translates these physical-consistency gains into a 37.4% relative improvement in task success over the strongest controlled internal baseline.

Kerui Li, Zhe Jing, Chenyi Huang +4

Institute of Automation, Chinese Academy of Sciences · GigaAI · Beijing Institute of Technology

Physical Consistency in Video Generation Robotic Manipulation Preference Optimization Synthetic Data Generation for Robotics

Oct 5, 2026cs.SD

AuraSE: Low-Hallucination Generative Speech Enhancement via Multimodal Flow Matching and Inference Policy Optimization

Generative speech enhancement models can produce cleaner and more natural-sounding speech than conventional discriminative approaches, but may hallucinate by changing speech content or speaker identity, even with transcript conditioning. We present AuraSE, a flow-matching framework that addresses hallucination through complementary modality and inference designs. First, a double-stream-to-single-stream multimodal Diffusion Transformer (MMDiT) allows transcript and acoustic representations to interact while preserving a dedicated pathway for the degraded input. Second, we find that the best decoder configuration, governed by guidance scale, sampling temperature, and step count, varies substantially across utterances. This observation motivates Inference Policy Optimization (IPO), an online, on-policy preference optimization method. IPO generates multiple candidates from the current model under different inference configurations, ranks them with a multi-objective reward, and learns from their relative preferences. AuraSE-IPO ranks first on 11 of 12 metrics across the synthetic test sets and obtains the highest DNSMOS and blind-listening scores among the evaluated systems on the real DNS blind test set. At deployment, it uses a fixed

10

-step ODE decoder without classifier-free guidance (CFG) or per-utterance configuration search.

Yingda Shen, Yao Qian, Yuxuan Hu +7

The Chinese University of Hong Kong, Shenzhen · Microsoft Research

Inference-Time Optimization Audio Diffusion Models Speech Enhancement Preference Optimization

Oct 2, 2026cs.SD

Rubric-Based Optimization for Text-to-Music Generation

Post-training text-to-music generation requires reward signals that capture multiple aspects of musical quality beyond what any single automatic metric can measure. We study structured, rubric-based rewards from pretrained audio-language models (ALMs) as training signals for both autoregressive and diffusion-based music generators. An ALM scores each generated clip against the rubric; we rank candidates generated for the same text prompt by their scores and convert these rankings into preference pairs for DPO on both MusicGen-small and ACE-Step v1, and additionally use the rubric scores directly as scalar rewards for DiffusionNFT on ACE-Step v1. On MusicCaps, rubric-based optimization improves CLAP, SongEval, and Audiobox-Aesthetics simultaneously, with the strongest gains obtained by DiffusionNFT on ACE-Step. By contrast, on MusicGen-small, building preferences from any one of these automatic evaluators produces clear cross-metric trade-offs: the targeted evaluator improves while other independent evaluators deteriorate. We further study tempo, key, and instrumentation, where precise objective rewards are available. Directly optimizing these specialized rewards reliably improves the target attributes, whereas ALM rubrics provide only partial transfer for tempo and instrumentation and no measurable improvement for key. Together, these results suggest a practical division of labor: ALM rubrics are effective for broad perceptual qualities that are difficult to formalize, while specialized objective rewards remain preferable when reliable measurements are available.

Ping Wang, Guang Yang, Shao-Rong Su +3

Paul G. Allen School of Computer Science & Engineering, University of Washington · Department of Electrical and Computer Engineering, University of Washington · Allen Institute for AI

Text-to-Music Generation Rubric-Based RL Preference Optimization Rubric-Guided Generation

Oct 1, 2026cs.AI

Mitigating Social Sycophancy via Pluralistic Preference Optimization

Personal advice, including relationship advice, now ranks among the most common uses of generative AI. But language models (LMs) exhibit sycophancy: they affirm users much more often than humans do, which can make people overconfident and less willing to repair their relationships after a conflict. Prior work on mitigating sycophancy has focused on factual settings where a response can be checked against a ground truth answer, while mitigations for social sycophancy (e.g., personal advice, where there is no ground truth) have relied on simple prompting and post-training methods with limited effectiveness. Our insight is that social sycophancy occurs in part because LMs overly center on the user and fail to consider the perspectives of other stakeholders impacted by the user's behavior. To address this problem we propose Pluralistic Preference Optimization (PlurPO): given inputs describing interpersonal conflicts, the LM identifies and simulates the relevant stakeholders, and is then trained to prefer and generate responses acceptable to all stakeholders. PlurPO uses only signals the model produces about its own outputs, without ground-truth labels. PlurPO substantially reduces social sycophancy across four datasets and four model families compared to prior methods. For example, on statements of intent to cause harm, where the users' actions should not be endorsed, PlurPO reduces the endorsement rate by 89% on average across four models. On general advice questions, where the target is to match the endorsement rate of human responses, it closes the gap by more than half, from 17.8% to 8.0% on average. The preference dataset constructed by PlurPO for an 8B model also effectively transfers to mitigating sycophancy in a larger (32B) model. Our results indicate that social sycophancy can be reduced by leveraging a model's own capabilities to simulate a plurality of relevant perspectives.

Stephane Hatgis-Kessell, Myra Cheng, Xiaoxuan Hou +4

Department of Computer Science, Stanford University · Amazon · Department of Computer Science, University of Washington

LLM Sycophancy Preference Optimization

Oct 1, 2026cs.CL

GAW-PO: Preference Optimization with Gradient-Aligned Token Weights

Most preference optimization methods, such as Direct Preference Optimization (DPO), apply preference supervision at the response level, although autoregressive language models are optimized token by token. As a result, all tokens in a rejected response contribute to the negative training signal, including tokens that may encode behavior that is useful for the preferred response. We introduce GAW-PO, a gradient-aligned token reweighting method for DPO that estimates, for each rejected token, whether penalizing it would interfere with the preferred update directions. Tokens whose gradients are strongly aligned with the preferred behavior receive a weaker negative contribution, while conflicting tokens retain a stronger penalty. Our method achieves the highest average performance among the evaluated preference-optimization methods, improving by 0.97 points over standard DPO and 0.65 points over the strongest competing baseline across 11 benchmarks spanning mathematics, reasoning, coding, and question answering. We further show that gradient-aligned weighting is substantially more robust to aggressive preference optimization: as the DPO regularization parameter

β

decreases, standard DPO degrades sharply, whereas GAW-PO continues to improve. These results suggest that accounting for the interaction between rejected-token updates and preferred behavior provides an effective form of token-level credit assignment for preference optimization.

Andreea Dutulescu, Stefan Ruseti, Mihai Masala +2

National University of Science and Technology POLITEHNICA Bucharest · NVIDIA

LLM Alignment Direct Preference Optimization Adaptive Loss Weighting Preference Optimization Preference-Based RL

Sep 29, 2026cs.LG

Uncertainty-Normalized Margins for Direct Preference Optimization

Direct preference optimization (DPO) models binary preferences through a Bradley-Terry model with a common noise scale, without explicitly accounting for preference strength or prompt-dependent uncertainty from human feedback. We introduce uncertainty-normalized margin DPO (UNM-DPO), which combines strength-dependent margins with a learned prompt scale. Motivated by a heteroskedastic Bradley-Terry model, we develop two training objectives. Both compare the implicit rewards of preferred and rejected responses, derived from response log-probability ratios to a reference policy. Advantage-only (AO) divides this reward difference by the prompt scale before subtracting the margin; whole-residual (WR) subtracts the margin before dividing by the scale. For the WR comparison model, we establish a necessary and sufficient condition under which known margins make the prompt scale identifiable. We introduce a practical procedure for learning the scale. Building on WR, we introduce ULNM-DPO-WR, which normalizes each response's implicit reward by its length. We evaluate our methods against DPO and related baselines on HelpSteer2 and HelpSteer3, using the Skywork reward model as a judge. With Llama-3.1-8B-Instruct, ULNM-DPO-WR achieves tie-adjusted win rates against matched DPO of 68.00% and 65.31% on evaluation panels, with higher mean rewards and shorter responses on average. On AlpacaEval with a GPT-4.1 judge and GPT-4-Turbo reference answers, the same 8B policy achieves a length-controlled win rate of 21.62%, compared with 16.39% for DPO and 15.30% for SimPO. These results demonstrate the potential of combining preference-strength margins, learned prompt scales, and length normalization for policy optimization.

Sadegh Khorasani, Petrus Mikkola, Matthias Grossglauser

School of Computer and Communication Sciences, EPFL, Lausanne, Switzerland · University of Helsinki, Helsinki, Finland

Pairwise Preference Learning Direct Preference Optimization Preference Optimization Preference Learning

Sep 28, 2026cs.CV

Preference-Guided Adaptation for Open-Vocabulary Semantic Segmentation via Prompt Disagreement

Open-vocabulary semantic segmentation (OVSS) enables pixel-level prediction over arbitrary text-specified vocabularies and has shown strong generalization on common benchmarks. However, OVSS performance often degrades in specialized domains such as medical imaging, remote sensing, and industrial inspection, where dense pixel-level masks for adaptation are costly to obtain and require domain-specific expertise. We propose a preference-guided adaptation framework that replaces dense mask supervision with binary preferences. We observe that different prompt templates produce systematically different segmentations for the same image, a phenomenon we call prompt disagreement, and we repurpose it as a built-in source of preference supervision. Building on this, we mine localized preference queries from regions of high cross-template uncertainty, and adapt the OVSS model with Region-Localized Preference Optimization (RLPO) together with consistency regularization that stabilizes updates outside the queried region. Across extensive experiments on the MESS benchmark, the proposed method achieves consistent gains across diverse OVSS backbones without any pixel-level annotation, and remains effective under noisy preferences. Our code is available at https://github.com/blue-531/pref-ovss.

Hyun-Kurl Jang, Jihun Kim, Kuk-Jin Yoon

Visual Intelligence Lab. KAIST

Weakly Supervised Semantic Segmentation Preference Optimization Open-Vocabulary Semantic Segmentation

Sep 27, 2026cs.AI

Agentic Multi-Turn Reasoning: A Fairness Approach

Recent advances in Large Language Models (LLMs) have enabled agentic systems capable of solving complex tasks through multi-turn planning, tool use, verification, and memory updates. However, learning agentic systems remains difficult due to two fundamental challenges, i.e., (1) long-horizon credit assignment, where supervision is available only at the final outcome, and (2) imbalanced data distributions, where dominant data patterns bias optimization and weaken adaptation to rare but informative reasoning behaviors. In this paper, we propose Fair Multi-Level Preference Optimization (Fair-MPO or

Φ

-MPO), a new preference optimization framework for agentic learning. We first show that Multi-Level Preference Optimization provides a principled and more computationally efficient framework for long-horizon reasoning. Then, we introduce a Fair Multi-Level Objective that addresses imbalance in agentic learning. We provide a comprehensive theoretical analysis demonstrating that our approach addresses both long-horizon reasoning and data imbalance. Our experiments on agentic reasoning benchmarks demonstrate that our approach achieves State-of-the-Art (SOTA) performance.

Thanh-Dat Truong, Sankalp Pandey, Hugh Churchill +3

CVIU Lab, University of Arkansas, USA · Dep. of Physics, University of Arkansas, USA · Dep. of Geosciences, University of Arkansas, USA +1

Preference Optimization Agentic Reasoning Long-Horizon LLM Agents

Sep 16, 2026cs.CL

A Zeroth-Order Paradigm for LLM Preference Alignment

Direct preference alignment methods are widely used to align large language models (LLMs) with human preferences because of their computational and memory efficiency. However, likelihood displacement motivates alternative ways to extract information from preference pairs with small likelihood margins. In this paper, we propose and analyze Comparison-based Preference Optimization (ComPO), a zeroth-order alignment method based on comparison oracles. ComPO extracts directional information from these pairs without directly optimizing a differentiable preference loss on them. We establish a convergence guarantee for its basic offline scheme under smoothness, gradient sparsity, and compatibility between the oracle and a latent objective. We further introduce online ComPO, which retains the offline comparison mechanism and uses unlabeled policy generations for reverse-KL control relative to a reference policy. Following the coverage perspective of preference fine-tuning, we establish a performance guarantee for a basic constrained scheme under local coverage and in-distribution pairwise reward accuracy. Experiments on Mistral, Llama, Gemma-2, Qwen3, and Gemma-3 models demonstrate improvements over existing direct alignment methods, including length-controlled win rates, with pair-level diagnostics providing evidence consistent with mitigating likelihood displacement.

Peter Chen, Xi Chen, Wotao Yin +1

Department of Electrical Engineering and Computer Sciences (EECS) University of California, Berkeley Berkeley, CA 94720, USA · Stern School of Business New York University New York, NY 10012, USA · Decision Intelligence Lab (Seattle) DAMO Academy, Alibaba Group U.S. Bellevue, WA 98004, USA +1

Pairwise Preference Learning LLM Alignment Preference Optimization

Sep 14, 2026cs.CL

StalePO: Anchored Token-Level Preference Optimization using Legacy Post-Edits in Machine Translation

Machine translation systems are periodically upgraded to stronger models, but the available preference signal is human post-edits of an older system's outputs, which the newer model may already surpass. Moreover, collecting fresh post-edits for every new model is prohibitively expensive. We call this the Stale Preference problem. Standard DPO can fail in this setting: it may increase the likelihood of inferior post-edits, erode the model's existing quality, and fail to provide the per-token control needed to correct localized errors. We introduce StalePO, an objective derived from three requirements this regime imposes. Likelihood movement must be downward on both responses, the policy must be anchored to its own base response, and the KL constraint must apply at the token level. These requirements are jointly necessary. In ablations, each mechanism in isolation leaves the model's performance indistinguishable from the base model, and only their combination converts stale feedback into gains. On English-to-Hindi and English-to-Turkish localization data, StalePO improves the fraction of segments passing all LLM-as-judge MQM quality checks by 14.9 and 4.6 percentage points, respectively, with gains concentrated on style and fluency. A human evaluation under the same framework confirms these gains on English-to-Hindi, raising the fraction of segments passing all seven human checks by 13.8 percentage points.

Rohit Dhaipule, Sukhdeep Singh Kharbanda, Prasanth Bathala +2

Translation Services, Amazon

Preference Optimization Language Model Post-Training Machine Translation

Sep 14, 2026cs.SD

Direct Preference Density Alignment for Conversational Audio Equalization

Large Language Model alignment typically relies on learned proxy reward models, which significantly increase the memory footprint during training and are notoriously prone to instability and reward hacking. While offline methods like Direct Preference Optimization (DPO) bypass the reward model, they lose the ability to perform online exploration. If no optimization constraints are applied, this can lead to format collapse in bounded, continuous spaces. To resolve this, we propose Direct Preference Density Alignment: An alternative framework that removes the need for a learned proxy reward model while strictly preserving the benefits of online reinforcement learning. We leverage large-scale user data (approximately 90,000 samples) to construct non-parametric preference density maps, establishing an empirical reward surface. In addition to removing the reward model, Direct Preference Density Alignment enables the combination of the online structural grounding of Group Relative Policy Optimization (GRPO) with the targeted offline refinement of DPO. We show that this GRPO+DPO combination achieves the highest performance, and in a blind audio equalization listening test, enables a 1.5B-parameter model to achieve perceptual parity with a carefully prompt-engineered GPT-4o mini baseline, using only a fraction of the inference compute.

Ioannis Stylianou, Sven Ewan Shepstone, Jon Francombe +2

Department of Electronic Systems Aalborg University Aalborg, Denmark · Bang & Olufsen A/S Struer, Denmark

RL for Language Models LLM Alignment Direct Preference Optimization Preference Optimization

Sep 9, 2026cs.CL

Direct Diversity Optimization for Diverse Successful Trajectories in Preference Post-Training

LLM agents for sequential decision tasks are often post-trained with trajectory-level outcome labels, but such labels provide little supervision for preserving multiple successful branches from the same decision state. We study this problem as successful strategy coverage: how broadly a model realizes distinct successful strategies under a fixed rollout budget. We present Direct Diversity Optimization (DDO), an offline post-training method that combines Divergence-Tree Collection (DTC) with the Reference-Relative Target-Odds Objective (RTO). DTC constructs state-aligned branch sets rooted at shared decision states, and RTO trains the model to match reference-relative targets over successful alternatives. DDO achieves the strongest task success and successful strategy coverage among the compared post-training methods across BabyAI, BabaIsAI, and WebShop. It also achieves the highest recovery rate after local action replacement and higher task success and coverage than successful-only imitation and decoding-time diversification controls.

Junwon Ko, Dong-Jae Lee, Minchan Kwon +2

School of Electrical Engineering, KAIST Daejeon, Republic of Korea

Offline RL Preference Optimization LLM Agent Training RL Post-Training

Sep 8, 2026cs.LG

Suan: Rectifying Direct Preference Safety Alignment in Large Language Models

Integrating robust safety guardrails into Large Language Models (LLMs) is essential for delivering helpful yet harmless responses. While proprietary systems exhibit reliable safety controls, their underlying methodologies and trade-offs remain largely undisclosed. Achieving comparable security in open-weight models remains a persistent challenge, as post-trained variants frequently suffer from over-refusal and degraded general quality. To overcome these drawbacks, we introduce Suan, a novel preference optimization algorithm. Unlike existing methods, we formulate the optimization objective directly at the gradient level, bypassing the standard variational derivation. As a result, we obtain more interpretable and robust training dynamics. Extensive evaluations across a diverse suite of competitive baselines and benchmarks demonstrate that Suan achieves superior safety alignment while fully preserving response utility.

Oleksandr Cherednichenko, Roman Klypa

Department of Mathematics and Mathematical Statistics Integrated Science Lab, Umeå University Umeå, Sweden · Univ. Grenoble Alpes, CNRS, Grenoble INP, LJK 38000 Grenoble, France

LLM Alignment LLM Safety Alignment Preference Optimization Multi-Objective Language Model Alignment

Sep 8, 2026cs.LG

TV-Regulated OPD: Direction Matters in On-Policy Distillation

On-Policy Distillation (OPD) facilitates the transfer of knowledge from domain expert to student in the post-training phase of Large Language Models (LLMs). However, the supervision signals in mainstream OPD methods suffer from high variance and noise which is generally instable during training. In this work, we systematically investigated what really matters to the performance and the fundamental mechanisms behind the instability during training. We found that retaining only the sign of token-level advantages is sufficient to achieve the performance comparable to standard OPD. Meanwhile, smoother and bounded advantages can stabilize the training process without sacrificing its performance. These motivated us to shape the advantages using the Total Variation (TV) and propose a robust TV regulated On-Policy Distillation (TV-OPD) method. Benefiting from the bounded and diminished advantages, TV-OPD exhibits stable training dynamics and steady late-stage performance. We conducted comprehensive experiments and found that, across various settings, TV-OPD consistently achieved better performance and lower variance in the late-stage of training.

Han Xiao, Yifan Niu, Dongyi Liu +2

The Hong Kong University of Science and Technology (Guangzhou) · The University of Edinburgh · The Hong Kong University of Science and Technology

Preference Optimization On-Policy Distillation RL Post-Training

Sep 8, 2026cs.LG

CUNO: Curriculum and Preference Optimization for Stable Graph Unlearning under Mass Deletion

Graph unlearning removes the influence of designated training data from a trained graph model without retraining from scratch. However, existing methods suffer a sharp drop in model utility under large deletion ratios (mass deletion), a phenomenon we refer to as catastrophic unlearning. We find that a key cause is the uniform treatment of all deleted samples, which is particularly damaging in graph learning: structural dependencies cause different nodes to play vastly different roles in the learned model, yet existing methods apply the same forgetting operation to the entire forget set. Based on this insight, we propose CUNO, a curriculum-based graph unlearning framework that removes the forget set progressively, ordering samples by their estimated unlearning difficulty across multiple stages. CUNO further employs a distribution-level negative preference optimization (NPO) objective at each curriculum stage that steers the model away from its original behavior on the current forget subset while preserving retained performance. Our theoretical analysis shows that the curriculum design is most beneficial when the forget set spans a wide range of unlearning difficulty, a condition naturally satisfied under mass deletion. Comprehensive experiments confirm that CUNO consistently mitigates catastrophic unlearning: at 20% deletion, it retains 74% of the original utility compared to 26-53% for existing methods, and maintains more than half the original utility even at 50% deletion. Our code is publicly available at https://anonymous.4open.science/r/cuno-D4FF.

Chenhan Zhang, Ali Braytee, Madhushi Bandara +4

University of Technology Sydney

Preference Optimization Machine Unlearning

Sep 6, 2026cs.AI

MARBO: Relational Belief Grounding for LLM Agents in Social Deduction Games

Social deduction games (SDGs) require agents to reason under partial observability by maintaining relational beliefs about hidden roles and team alignments. While recent LLM-agent approaches improve gameplay through prompting and preference optimization, they often optimize actions and in-game speech without explicitly grounding them in such beliefs. This frequently leads to strategically inconsistent behavior, especially for compact LLM agents. We introduce Multi-Agent Relational Belief Optimization (MARBO), a belief-grounded preference optimization framework that leverages relational beliefs to guide strategic decisions and in-game speech. MARBO provides preference feedback only when behaviors are supported by reliable relational beliefs and lead to strategically favorable social outcomes, encouraging more consistent learning under uncertainty. Experiments on representative SDGs show that MARBO enables compact LLM agents to consistently outperform existing baselines. The Code is available on https://github.com/PleaseTakemeAway/MARBO.

Yechan Hwang, Sangjun Bae, Jeongmo Kim +2

Graduate School of Artificial Intelligence, UNIST, Ulsan, South Korea.

Game-Playing Agents Social Deduction Games Preference Optimization LLM Agent Training Multi-Agent Communication

Sep 4, 2026cs.CV

MCPO: Modality-Contrastive Preference Optimization for Multimodal Chain-of-Thought Compression

Recently, multimodal large-scale reasoning models have demonstrated remarkable capabilities in solving complex tasks through long Chains-of-Thought (M-CoT). However, excessively long reasoning trajectories incur substantial computational costs and significant KV-cache pressure. Existing CoT compression and alignment paradigms mainly rely on static rules or single-dimensional preferences, lacking fine-grained cross-modal constraints; as a result, they are prone to inducing visual laziness and hallucinatory reasoning. To address these issues, we propose Modality-Contrastive Preference Optimization (MCPO), a highly sample-efficient two-stage length-compression method that requires fewer than 900 training samples. In the compression stage, we introduce a step-level Normalized Cross-Modal Mutual Information (NCMI) pruning algorithm, which automatically identifies and removes visual-independent reasoning steps by comparing the reasoning discrepancies between with-image and no-image contexts. This significantly reduces redundancy and hallucinatory content in the reasoning chains. In the alignment stage, the model first undergoes supervised fine-tuning to achieve domain-adaptive initialization, followed by optimization using an asymmetric multimodal length-controlled preference loss. This objective adopts a highly nonlinear odds-ratio formulation that provides steep gradients in the with-image context to reinforce length constraints for preferred trajectories, while applying a scaled, flat-gradient linear difference in the no-image context to maintain modality consistency, thereby achieving stable cross-modal preference alignment. Extensive experiments on mainstream base models such as Qwen3-VL-Thinking show that our method can reduce CoT length by up to 69.5% and achieve up to 3.34x end-to-end inference speedup while preserving original accuracy.

Guangheng Yang, Zhenliang Ni, Zhenkai Wu +4

Huawei Technologies Ltd. Shanghai, China

Efficient Multimodal Inference Multimodal Large Language Models Preference Optimization Multimodal CoT Reasoning

Sep 3, 2026cs.AI

FiMI Banking: A Sovereign Model for Indian Retail Banking

Banks need conversational systems that can answer product questions, assist customers with account-related requests, and operate safely within strict operational and regulatory constraints. General-purpose language models do not reliably meet these requirements. They fall short when a task requires grounded information, correct tool use, or cautious handling of bank-specific sensitive situations. We introduce FiMI Banking, a controlled Indian retail-banking setting. We build it from vetted banking documents, structured ground truth, synthetic customer backgrounds, and banking tools. We evaluate two post-training approaches: preference optimization for response-level behavior, and reinforcement learning with verifiable rewards for multi-turn tool-use tasks. Preference optimization improves safe behavior substantially: out-of-scope refusal rises from 52% to 80%. Reinforcement learning improves edge-case performance from 0.509 to 0.718 and order-sensitive task performance from 0.590 to 0.679, while using 29% fewer generated tokens. These results show that preference optimization and verifiable-reward reinforcement learning address complementary requirements for reliable banking agents.

NPCI AI Research Team, Aman Kumar, Asit Desai +15

NPCI AI Research Team

Financial Services RL for Language Models Preference Optimization Reinforcement Learning with Verifiable Rewards AI Agent Safety Benchmarks LLM Agent Reliability

Sep 2, 2026cs.AI

CoMerge: Conflict-Driven Preference Optimization for Multi-Task Model Merging

Model merging provides an efficient paradigm for constructing multi-task large language models (LLMs) without full model retraining, yet it remains challenged by parameter interference. While existing methods aim to preserve the capabilities of individual expert models and mitigate interference, they generally do not directly learn from the potentially degraded behaviors exposed by naive merging. In this paper, we propose a conflict-driven preference optimization framework for model merging (CoMerge), which reformulates model merging as a preference optimization problem. The approach utilizes a self-supervised, conflict-driven strategy that leverages the defects of naive merging methods (e.g., task arithmetic) as hard negative samples to construct preference pairs without external annotations. By applying preference optimization to refine lightweight, tensor-wise merging coefficients, CoMerge enables the model to mitigate parameter-space conflicts while preserving task-specific capabilities. Extensive experiments show that CoMerge achieves an average normalized performance of 0.9968 on MergeBench, outperforming all evaluated data-free and data-driven model-merging baselines. Furthermore, on Llama-3.1-8B-Instruct, CoMerge yields marked improvements on conflict-sensitive tasks such as instruction following and safety, while remaining highly competitive with full-parameter fine-tuning despite optimizing only 1,445 scalar coefficients.

Mingjie Zheng, Zihao Chen, Wenqing Chen +4

School of Software Engineering, Sun Yat-sen University, Zhuhai, China · The State Key Laboratory of Blockchain and Data Security, Zhejiang University, Hangzhou, China · School of Artificial Intelligence, Sun Yat-sen University, Zhuhai, China +1

Model Merging Preference Optimization

Sep 1, 2026cs.CL

Ready to Speak: Aligning LLMs for TTS-Friendly Text Generation

Current Large Language Models (LLMs) are primarily optimized for written text, often producing outputs that are grammatically correct and helpful yet poorly suited for spoken delivery via Text-to-Speech (TTS). In this work, we study how to make LLMs natively generate TTS-friendly text, which we frame as a preference alignment problem: instead of relying on downstream rewriting modules, we directly align LLMs to generate text optimized for spoken delivery. We introduce two preference datasets spanning different target domains, CORA and Recipe, which contain paired TTS-friendly and TTS-unfriendly responses. We further propose an evaluation suite combining a pattern-based heuristic metric, a TTS

\to

ASR evaluation pipeline, and a MUSHRA listening study with human judges. Our experiments compare the recently proposed Feature-aware Sampling and Tuning (FaST) framework -- leveraging interpretable features instead of a black-box reward model -- against an array of alignment baselines on the TTS-friendly generation task. Notably, we found that FaST achieves the best overall tradeoff between TTS-friendliness and helpfulness across various settings. We also identified a strong correlation between our different metrics, highlighting the ability to reliably assess TTS-friendliness via an efficient heuristic.

Thibaut Thonet, Jos Rozen, Laurent Besacier

NAVER LABS Europe

LLM Alignment TTS Evaluation Preference Optimization Text Generation

Aug 31, 2026cs.LG

Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization

Sycophantic agreement refers to a behavior in which language models excessively affirm the user, often at the cost of factual accuracy. Although sycophantic agreement is a well-known failure of model alignment, there is limited understanding of how it emerges from model training. In this work, we demonstrate that sycophantic agreement can emerge as an unintended consequence of widely used contrastive preference optimization objectives. Using the OLMo 3 post-training pipeline, we show that, for various pairs of teacher models across three families, there is a strong correlation between the log-ratio of the teacher model sycophantic agreement rates and the resulting student model sycophantic agreement rate. We further demonstrate that this unintended transfer is not limited to DPO but also occurs across 6 other preference optimization objectives. To understand whether this effect can be attributed to particular training examples, we analyze the preference data and find that the sycophancy signal is diffused across the entire dataset rather than concentrated in a sparse set of examples: each example appears neutral, i.e., there are no explicit instances of sycophantic agreement, and filtering based on probe-based data attribution or logit-linear selection fails to mitigate sycophancy without removing a large portion of the dataset. Overall, our findings suggest that the teacher models used to generate preference data can interact with alignment training objectives in unexpected ways, generalizing to undesirable and potentially harmful behaviors like sycophantic agreement.

Camila Blank, Zhuofan Ying, Christopher Potts +2

1Stanford University · 2Columbia University

LLM Sycophancy Emergent Misalignment in Language Models Preference Optimization Preference Alignment

Aug 31, 2026cs.CR

Balancing Privacy, Utility, and Safety in LLM Alignment through Preference Optimization

Preference optimization is widely used to align large language models with human preferences, but preference-data composition may also influence privacy-relevant memorization. We examine whether adding synthetic privacy-preference pairs to Direct Preference Optimization (DPO) is associated with lower canary-based memorization signals without modifying the objective or introducing a formal privacy mechanism. We propose Privacy-Pressure Preference Mixing (P3M), a data-composition protocol that varies the amount of privacy-preference data while keeping helpfulness and harmlessness preference data fixed. We evaluate a non-privacy Baseline and privacy-mixing ratios of 0.5, 1.0, and 2.0 using Gemma 3 270M-IT across five random seeds and validate the same four conditions using 4-bit-quantized Gemma 2 2B-IT across three seeds. Overall, under the tested conditions, privacy-preference mixing is associated with lower mean canary suffix log-likelihood proxy values across both model settings and lower aggregate membership-inference attack performance relative to the Baseline in the mixed-source 2B evaluation. Specifically, across the privacy-aware 2B configurations, the mean area under the receiver operating characteristic curve (AUROC) ranges from 0.596 to 0.629, and the mean area under the precision-recall curve (AUPRC) ranges from 0.541 to 0.575, compared with 0.804 and 0.790, respectively, for the Baseline. However, the reduction in membership distinguishability does not hold uniformly across data sources. Moreover, the relationship between the privacy ratio and harmlessness preference accuracy varies by model setting, whereas helpfulness preference accuracy remains broadly stable. These findings suggest that P3M should be viewed as a lightweight empirical protocol for examining privacy-utility-safety trade-offs rather than as a formal privacy guarantee or a defense against extraction attacks.

Dishu Yang, Jingjing Liu, Jize Li

Khoury College of Computer Sciences Northeastern University San Jose, USA · Department of Electrical Engineering and Computer Sciences University of California, Berkeley Berkeley, USA · Department of Economics Boston University Boston, USA

LLM Alignment Memorization in Language Models LLM Safety Alignment Privacy Leakage in Language Models Preference Optimization Privacy-Utility Tradeoff

Aug 13, 2026cs.SD

VoxAudio: Vocalized Audio Synthesis via Multi-Reward Autoregressive Flow Matching

Vocalized audio synthesis, the task of generating audio in which intelligible speech is embedded within an environmental soundscape, underpins applications such as podcast production and video dubbing. Existing Text-to-Audio (T2A) systems either reduce quoted speech to unintelligible vocal murmur or delegate it to a separate TTS model with post-hoc mixing, which forfeits control over when speech occurs and how it interacts with the scene. We present VoxAudio, a causal autoregressive flow matching model that addresses this problem from three complementary aspects. At the architecture level, chunk-wise causal factorization with independent per-chunk noise levels lets audio be emitted through sliding-window streaming inference with KV caching at variable target durations; to enable inference at arbitrary chunk granularities, we further pretrain the model with randomized chunk boundaries. At the preference level, multi-reward Negative-aware FineTuning (NFT) jointly optimizes semantic fidelity, linguistic accuracy, aesthetic quality, and temporal grounding At the data level, to supply the missing supervision for vocal content, we build VoxCorpus, a large-scale corpus whose captions quote the verbatim transcript of embedded speech with time intervals, and VoxBench, an interval-annotated benchmark with a temporal-grounding metric. Experiments on four benchmarks spanning general audio, speech, and unified vocalized audio validate the effectiveness and efficiency of VoxAudio. Our code and demos are available at https://voxaudio.github.io.

Wenxiang Guo, Changhao Pan, Ziyue Jiang +2

Zhejiang University, Hangzhou, China

Text-to-Audio Generation Preference Optimization Audio Generation

Aug 12, 2026stat.ML

SSPO: Structure-Aware Similarity-Weighted Preference Optimization for Neural Combinatorial Optimization

Neural combinatorial optimization (NCO) relies on parallel solution sampling for training, yet existing methods fail to fully exploit the rich information latent in a co-sampled solution group. Preference-optimization methods anchor on the single best solution and discard fine-grained quality and structural signal from all other peers-a failure we term gradient signal polarization. Mean-based baselines instead weight peers uniformly, so structurally near-identical peers flood the baseline with redundant information and keep gradient variance high-a failure we term baseline redundancy. We propose SSPO (Structure-Aware Similarity-Weighted Preference Optimization), which scores all

B

sampled solutions jointly through a dissimilarity-weighted leave-one-out baseline: structurally distinct peers receive higher weight, resolving both failures in a single mechanism. The baseline uses zero-parameter, problem-adaptive solution embeddings built from the encoder's existing node representations. Experiments on TSP, EFL, and JSP benchmarks show consistent gains over prior best-anchor and uniform-weight baselines. A direct comparison against uniform RLOO on TSP and EFL confirms that structure-aware weighting is the primary driver of improvement. The SSPO-trained EFL policy has been deployed in a production facility-location system at JD

\mathord{.}

com, confirming practical viability at scale.

Yuanyu Li, Jintao Xu, Zijiang Liu +6

Supply Chain Tech Team Y, JD.com, Beijing, China · Department of Industrial Engineering, Tsinghua University, Beijing, China · Faculty of Engineering and Faculty of Business and Economics, The University of Hong Kong, Hong Kong, China

Combinatorial Optimization Preference Optimization Importance Weighting RL for Combinatorial Optimization Neural Combinatorial Optimization

Aug 10, 2026cs.IR

TSPORec: Token Selection via Preference Optimization for LLM-Based Sequential Recommendation

Large Language Models (LLMs) have emerged as powerful tools for improving recommendation systems. The effectiveness of LLMs arises from their ability to harness rich textual information and their capacity to model heterogeneous user preferences based on users' interaction history. However, due to the large-scale and deep architectures, LLM-based sequential recommendation approaches generally incur high inference costs, resulting in a low return on investment. To mitigate this cost, many existing approaches resort to using only the first few tokens of item descriptions, which inadvertently discards valuable information contained in the full text, thereby leading to suboptimal recommendation performance. To address this limitation, we propose a novel Token Selection approach for Preference Optimization in LLM-based sequential Recommendation, i.e., TSPORec, which accurately pinpoints informative tokens throughout the entire textual content to improve recommendation performance. Specifically, we design a three-stage pipeline to select informative tokens and introduce a novel proxy reward to facilitate the implementation. TSPORec not only enhances recommendation performance but also improves computational efficiency. Extensive experiments across two models and datasets demonstrate the superb performance (up to 31.25%) and efficiency (up to 63.4%) of our approach compared with six baseline approaches. Code is available at https://github.com/WNQzhu/TSPORec.git.

Wenqiao Zhu, Chao Xu, Haipang Wu +1

LLM Inference Efficiency Sequential Recommendation Preference Optimization Large Language Model-Based Recommendation

Aug 4, 2026cs.CL

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 Disagreement Preference Optimization Multi-Agent LLMs Preference Learning

Aug 3, 2026cs.RO

Adaptive Human-Robot Collaborative Painting Combining Preference-Based Optimization and Dynamic Motion Primitives

This work presents a human-centered collaborative framework that integrates Preference-Based Optimization (PBO) and Dynamic Movement Primitives (DMPs) to optimize robot-assisted tasks such as painting. The system allows the operator to perform the process while the robot adapts its behavior in real-time, dynamically adjusting the orientation of the piece in order to match the orientation of the operator's hand. The PBO framework leverages the GLISp algorithm to iteratively refine control parameters such as execution time, robot responsiveness, and rotation amplification through human feedback. Moreover, DMPs have been modified to enhance the reactive behavior of the robot and its adaptability to ergonomic requirements. The method was validated with a heterogeneous group of participants executing \rev{painting tasks}. The results show that our strategy effectively reduces operator effort while optimizing process outcomes.

C. Cella, M. Ristic, M. Faroni +2

Politecnico di Milano

Robot Motion Generation Human-Robot Collaboration Robotic Manipulation Preference Optimization

Aug 2, 2026cs.CL

Cloud-ScPO: Hidden-State Geometry for Semi-Supervised Preference Optimization in LLM Reasoning

Preference optimization improves mathematical reasoning in large language models (LLMs), but reliable chosen-rejected pairs usually require verified answers, human annotations, or external reward models. We investigate whether preference supervision can instead be derived from the model's internal representation geometry in a semi-supervised setting. Our analysis shows that reasoning trajectories generated across different mathematical problems form structured global point clouds in which correct and incorrect trajectories exhibit different geometric organization. Based on this observation, we propose Cloud--ScPO, a topology-guided preference-mining framework that uses a small labeled set to construct multiple correct and incorrect reference Clouds. Each trajectory is represented by a mean-pooled hidden state and scored against connectivity-induced components using a component-level soft

k

-nearest-neighbor measure averaged across reference banks. We combine this cross-problem Cloud signal with prompt-level self-consistency: self-consistency determines the answer-level preference direction, while Cloud scoring selects concrete trajectories and filters pairs by their score margin. Experiments on GSM8K and MATH-Numeric across four model settings show that Cloud--ScPO consistently improves over ScPO, with gains of up to 4.49% on GSM8K and 4.19% on MATH-Numeric. Pair-level analyses further show that Cloud--ScPO maintains comparable correctness reliability while more effectively separating informative chosen trajectories from incomplete, repetitive, or otherwise low-quality rejected responses.

Yuzhou Liu, Xiyang Hu

University of Southern California · Arizona State University

Preference Optimization Semi-Supervised Learning Representation Geometry in Language Models LLM Mathematical Reasoning

Jul 30, 2026cs.CL

Rolling With Resistance: Preference-Optimized LLM Counselors Can Trade Goal Persistence for Relational Attunement in Motivational Interviewing

In Motivational Interviewing (MI), a client's sustain talk (arguments for the status quo) calls for the counselor to roll with resistance, a move that can fail in two opposite ways: capitulation (abandoning the change agenda to preserve rapport) or confrontation (arguing or directing, overriding the client's autonomy). We introduce a two-axis evaluation of counselor responses, anchored in the Motivational Interviewing Treatment Integrity (MITI) code, Goal Persistence (GP) and Relational Attunement (RA), yielding a four-quadrant framing in which rolling with resistance is high on both, and we ask whether penalizing one failure through preference optimization teaches rolling with resistance or provokes its opposite. From the expert-annotated AnnoMI corpus we build topic-disjoint Direct Preference Optimization data whose preference sets differ only in which failure is rejected, using on-policy negatives. An automatic judge, validated against AnnoMI's expert labels and rechecked by trained human coders, scores blind pairwise win-rates against each base under a firewall in which disjoint model families generate, label, and judge. Across three aligned instruction models spanning the Qwen and Llama families, penalizing confrontation reliably lowers goal persistence below parity, on every base and in every seed run, a robust cost, whereas the attunement gain is base-dependent, present on two of the three bases but absent on the third. Penalizing capitulation is inert, because these models rarely capitulate on-policy, so the trade is gated by each base's failure profile. A prompt-only control raises attunement without the goal-persistence cost, locating the cost in the optimization rather than in attunement itself.

Weiying Chen, Junlong Shen, Zhexuan Tang

University of Alberta · University of Shanghai for Science and Technology

Direct Preference Optimization Preference Optimization

Jul 30, 2026cs.CV

Temporal Concentration from Rollout Errors: Implicit Preference Optimization for Text-to-Video Diffusion

Recent advances in preference alignment for diffusion-based video generation, particularly via Direct Preference Optimization (DPO), have significantly improved visual quality. However, temporally sparse artifacts such as motion collapse, object flickering, and color oversaturation remain a major barrier to perceptual realism. Existing methods struggle with these issues due to two key limitations: (1) the preference attribution bottleneck, where offline human annotations are costly and fail to accurately capture learning dynamics, while online reward signals are rollout-aware but often unstable and biased; and (2) temporal credit misallocation, where uniformly applied supervision cannot effectively target the brief segments in which artifacts occur. To address these challenges, we propose concentrated Implicit Preference Optimization (cIPO), a post-training framework for video diffusion models. cIPO derives implicit preference signals directly from the denoising process: given a real video, the model adds forward noise and reconstructs it via iterative denoising, treating the original as the preferred sample and the reconstruction as the dispreferred one. This formulation captures inference-time errors without requiring human annotations or external reward models. Moreover, frame-level discrepancies between original and reconstructed videos reveal when failures occur. cIPO leverages this by computing temporal reconstruction errors and concentrating optimization on high-error segments, enabling more precise correction of failure-prone regions. Extensive experiments demonstrate that cIPO consistently enhances video authenticity and temporal coherence across multiple datasets, highlighting the effectiveness and efficiency of implicit preference with temporally concentrated optimization.

Henglin Liu, Fangyuan Kong, Jing Wang +7

Tsinghua University · Kling Team, Kuaishou Technology · Sun Yat-sen University

Diffusion Model Alignment Video Diffusion Models Text-to-Video Generation Temporal Credit Assignment Preference Optimization Temporal Consistency in Video Generation

Jul 30, 2026cs.LG

AutoPref: Automatic Discovery of Task-Specific Preference Objectives for Neural Combinatorial Optimization

Combinatorial optimization problems (COPs) underpin many real-world decisions, but their exponentially large search spaces make high-quality solutions costly to obtain. Neural combinatorial optimization (NCO) learns fast construction policies, typically with reinforcement learning (RL), while preference-based NCO improves sample efficiency by learning from relative solution quality. However, existing preference objectives combine two distinct design choices in manually specified, one-size-fits-all formulations: what learning signal to extract from each solution pair and how to weight each pair relative to the sampled set. We present AutoPref, the first LLM-guided framework for automated preference-objective discovery in NCO. AutoPref factorizes the objective into a pairwise loss program, which defines the learning signal, and a set-aware weighting program, which determines each pair's relative contribution. Their composition forms a unified programmatic objective space containing existing preference objectives as special cases. To make its search tractable, we introduce a staged conditional search strategy with behavioral gates that filter inadmissible programs before short-horizon training and evaluation. Across TSP, CVRP, FFSP, and JSSP, AutoPref consistently outperforms strong hand-designed baselines across problem scales, demonstrating the benefits and scalability of automated objective discovery for NCO.

Shengda Gu, Kai Li, Xinyi Ke +3

C2DL, Institute of Automation, Chinese Academy of Sciences · School of Artificial Intelligence, University of Chinese Academy of Sciences · Tencent AI Lab +1

Automated Algorithm Discovery Pairwise Preference Evaluation Preference Optimization Neural Combinatorial Optimization Preference-Based RL

Jul 29, 2026cs.CL

BridgeAlign: Bridging Preference Alignment for Humanities and Social Sciences

While data synthesis for large language models (LLMs) is prevalent, it primarily targets domains with verifiable answers, overlooking open-ended humanities and social sciences (HSS), where nuanced quality judgments matter more than objective correctness. This makes preference alignment a natural paradigm for broad HSS tasks. Yet existing methods are either costly or not tailored to broad HSS disciplines. We thus propose BridgeAlign, among the first preference-alignment pipelines for broad HSS disciplines, with three phases: i) Seed Curation: curating HSS seed documents from web corpora via heuristic/LLM-based filtering and text refinement; ii) Preference Data Synthesis: generating preference triplets via persona-based instruction inversion with Q&A consistency checks; iii) Preference Optimization: moving beyond naive human-vs-model heuristics by first grounding preferences in HSS quality rubric, then generating transitional responses via controlled quality degradation to form near-boundary preference pairs for finer-grained quality discrimination. Aligning over 210k synthetic preference samples, BridgeAlign enables Qwen3-8B to achieve the best average across 17 benchmarks against 11 strong baselines; importantly, leading on both human-preference and knowledge-based capabilities at once, with no trade-off between them, as supported by extensive experiments and contextualized by existing theories.

Ru Peng, Haokai Xu, Xijun Gu +11

Zhejiang University · Inclusion AI, Ant Group · Qwen Team, Alibaba Group +2

Pairwise Preference Learning Preference Optimization Synthetic Data Generation for Language Models Preference Alignment

Jul 29, 2026cs.CL

TELLER: Dual-Path Iterative Preference Optimization for Table Entity Linking

Entity linking in tables matches short and ambiguous cell mentions to their corresponding knowledge-base entities. Existing approaches typically rely on data preprocessing pipelines that retain either compact or extensive table content as contextual evidence, and then formulate entity linking as a language generation task for instruction-tuned models; recent systems further incorporate explicit reasoning to disambiguate challenging mentions. However, their training supervision is usually static: fixed preference data cannot adapt to the residual errors of an evolving model, while variations in reasoning length can bias sequence-level preference learning. To address these limitations, we present TELLER: Table Entity Linking through Learning from Errors and Reasoning. We first retrieve and rank Wikidata candidates and retain reduced table evidence in the prompt. The direct-answer path applies iterative direct preference optimization and refreshes its preference data with residual errors from the updated model. The reasoning path uses filtered and compressed chain-of-thought rationales for supervised fine-tuning, followed by our iterative length-normalized regularized preference optimization. On the TableInstruct entity-linking subset, the direct-answer path improves accuracy from 94.35% to 94.50%; on the MammoTab V2 evaluation set, it improves accuracy from 87.59% to 88.20%. The reasoning path improves accuracy from 92.90% to 92.95% on TableInstruct and from 79.09% to 81.85% on MammoTab V2, while maintaining high rates of complete reasoning generation. These results show that iterative preference learning benefits both concise entity prediction and explicit reasoning.

Yixin Peng, Kehao Li, Stefan Decker

RWTH Aachen, Ahornstraße 55, 52074 Aachen, Germany

Direct Preference Optimization Preference Optimization LLM Information Extraction Entity Linking

Jul 27, 2026cs.AI

Less Data, Better Alignment: Data-Centric Multi-Evaluator Agreement for Preference Optimization

Research on preference optimization often varies the training objective while holding the data fixed. We instead ask whether a small, high-confidence set of on-policy responses can provide a reliable learning signal. Our method, DMAPO (Data-centric Multi-evaluator Agreement for Preference Optimization), generates candidate responses from the target policy, evaluates helpfulness, factuality, and conciseness with rubric-specialized evaluators, applies a process-critic correction, and retains only high-consensus desirable or undesirable examples. This procedure accepts 1,871 of 54,236 Mistral-7B candidates (3.45%). KTO trained on this set reaches 7.50 on MT-Bench, 95.5% length-controlled win rate against a text-davinci-003 reference, and 57.3% IFEval prompt accuracy. Independent pairwise evaluation also favors DMAPO over SimPO: GPT-4o yields a net win rate of 23.3 points on 129 held-out prompts and 24.0 points on 200 out-of-distribution LMSYS-Chat prompts; Claude Opus 4.7 yields 24.1 points on the held-out set. Changing the evaluator model or rubric alters the selected examples but has little effect on downstream performance. A second-backbone study yields a similar 3.41% acceptance rate, although its performance gains are more modest. Across these experiments, consensus filtering offers a data-efficient route to preference optimization for general instructions, at the cost of additional curation compute and dependence on evaluator judgments.

Zhengtao Yao, Runhao Li, Xupeng Chen +12

University of Southern California · New York University · Columbia University +4

Preference Optimization

Jul 22, 2026cs.AI

EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference Optimization

Large Reasoning Models (LRMs) often suffer from overthinking due to redundant verification steps. Existing approaches for mitigating overthinking, such as fast-slow thinking switching and reasoning trajectory compression, fail to make a fine-grained distinction between beneficial and redundant steps within the LRM's reasoning process, and may thus impair reasoning capability in their pursuit of efficiency. To simultaneously improve reasoning efficiency and capability, we propose EvoThink, a framework that reduces redundant verification and encourages the exploration of new reasoning paths. EvoThink comprises two key components: Self-Pruning Training (SPT), an unsupervised method that iteratively prunes redundant reasoning steps and self-trains on the concise trajectories; and Aha-Moment Preference Optimization (AMPO), which, inspired by genetic algorithms, identifies valuable failed reasoning attempts, synthesizes from-wrong-to-right aha-moment data, and optimizes the model to internalize this reasoning pattern. Extensive evaluations across mathematical reasoning and code generation benchmarks demonstrate that EvoThink not only substantially reduces inference-time token usage but also improves the reasoning capability of LRMs.

Xinbang Dai, Zheyu Xin, Huikang Hu +7

Southeast University · Noah’s Ark Lab

Overthinking in Language Models Self-Training for Language Models Efficient Language Model Reasoning Preference Optimization Efficient Language Model Inference Large Reasoning Models

Jul 17, 2026cs.CL

RIMS: Preference Optimization via Smoothed Multi-pair Aggregation for Small-Scale LLM Retrieval-Augmented Generation

Small-scale language models (SLMs) are attractive for retrieval-augmented generation (RAG) in resource-constrained settings, but their limited capacity makes them highly sensitive to noisy or spurious retrieved evidence. Existing preference-based methods such as RoseRAG select only the hardest single preference pair via hard argmin/argmax, discarding the remaining signal; others treat multiple pairs as independent binary comparisons, resulting in low data utilization. We propose RIMS, a three-stage preference optimization framework comprising (1) synthetic chain-of-thought preference data generation via rejection sampling using the target SLM itself without relying on proprietary models, (2) a differentiable soft aggregation mechanism that replaces hard selection with a smooth operator, preserving gradient signal from all preference pairs while retaining the discriminative structure of margin-aware selection, and (3) preference optimization with the smoothed objective applied to multiple alignment algorithms. We theoretically show that the smoothed approximation admits a controllable error bound and that smooth aggregation yields provably tighter gradient alignment to the oracle objective than hard selection. Experiments on four multi-hop question answering benchmarks show that our approach outperforms state-of-the-art baselines across multiple SLM backbones, achieving consistent gains in Exact Match and F1 under noisy retrieval conditions. Our implementation is available at https://github.com/tptrix29/RIMS.

Pei Tian, Zihan Dong, Tianci Liu +2

Department of Computer Science, Columbia University · Department of Statistics, Rutgers University · School of Electrical and Computer Engineering, Purdue University +1

Retrieval-Augmented Generation Small Language Models Preference Optimization

Jul 17, 2026cs.LG

TD-DPO: Difference-Aware Preference Optimization for Mitigating Sycophancy in Clinical Autism Intervention Dialogue

The sycophancy of large language models can increase the safety risk in intervention dialogue for autistic children. Supervised fine-tuning can somewhat reduce sycophancy, but relying solely on positive examples is often insufficient to identify and correct failure patterns. We observe that sycophancy behaviors can often be localized to a limited span within the model response. In this regime, sequence-level preference optimization can over-update preference-irrelevant tokens and degrade intervention ability. To address this, we propose the \textbf{M}inimal \textbf{E}dit \textbf{D}ata \textbf{A}ugmentation (MEDA) strategy to construct controlled, stable, minimal edit preference pairs and \textbf{T}oken-level \textbf{D}ifference \textbf{D}irect \textbf{P}reference \textbf{O}ptimization (TD-DPO), which upweights difference tokens between chosen and rejected responses while downweighting shared tokens to suppress background drift. Extensive experiments across multiple backbones and evaluators show that TD-DPO achieves a better trade-off between sycophancy mitigation and intervention ability retention in our offline settings, highlighting its potential as a practical alignment approach for autism intervention.

Shuzhong Lai, Junhong Lai, Chenxi Li +5

1Nanhu Brain-Computer Interface Institute · 2MOE Frontiers Science Center for Brain and Brain-Machine Integration, Zhejiang University · College of Computer Science and Technology, Zhejiang University +3

LLM Sycophancy Direct Preference Optimization LLM Safety Alignment Preference Optimization

Jul 15, 2026cs.CV

Groc-PO: Grounded Context Preference Optimization for Truthful Multimodal LLMs

Despite the rapid progress of Multimodal Large Language Models (MLLMs), they still suffer from untruthfulness issues, such as visual hallucinations, content fabrication, and unfaithful reasoning, which substantially undermine their faithfulness and practical utility. Alignment methods based on human preference, such as Direct Preference Optimization (DPO), have been widely adopted to address these issues. However, multimodal reasoning errors often propagate across stages, and final-answer errors can often be traced to mistakes in early grounding stages, yet standard DPO typically applies preference optimization at the final-answer level. This credit-assignment challenge means that supervision for early grounding stages is indirect rather than stage-specific, making it difficult to suppress error propagation arising from grounding drift and context inconsistency. To address this, we propose Grounded Context Preference Optimization (Groc-PO), a grounded preference optimization framework for MLLMs. We further construct the Grounded Context Preference Dataset (GCPD), organizing multi-stage preference samples around three stages of Object Grounding, Contextual Grounding, and Grounded Reasoning, to capture the formation, integration, and utilization of grounded context. By introducing more explicit preference supervision over multiple grounded stages, Groc-PO strengthens context-dependent reasoning and mitigates cross-stage error propagation. Extensive experiments show that, compared with standard DPO and other strong baselines, Groc-PO achieves improved performance in hallucination mitigation, faithful reasoning, and overall reliability, supporting the value of more explicit grounded supervision for trustworthy multimodal reasoning.

Zhixiao Zheng, Zheren Fu, Zhiyuan Yao +3

University of Science and Technology of China Hefei, China · Xiaomi Corporation Beijing, China · State Key Laboratory of Communication Content Cognition, People’s Daily Online Beijing, China

Multimodal Grounding Multimodal Large Language Models Multimodal Reasoning Preference Optimization VLM Hallucination Mitigation

Jul 15, 2026cs.CL

Exploring Post-Training Alignment of Small Language Models for Biomedical Data-to-Text Generation: A Case Study of Medication Leaflet

Translating complex biomedical data into patient-friendly narratives is central to modern biomedical informatics. This study presents a comparative analysis of training small language models (SLMs) in specialized biomedical datato-text generation tasks. We explore widely adopted post-training methods including supervised fine-tuning (SFT), direct preference optimization (DPO), odds ratio preference optimization (ORPO), and group relative policy optimization (GRPO) with Qwen-based SLMs on a medicine package leaflets dataset. To assess cross-dataset generalizability, we also curated drug label data from openFDA. We evaluate models using both standard lexical overlap metrics like ROUGE as well as semantic similarity measures. Across our experiments, the results show that (1) the aligned SLMs outperform proprietary models like GPT-5; (2) ORPO outperforms the SFTbaselines; (3) GRPO yields the most robust cross-dataset performance among the alignment methods tested as well as GPT-5.

Xi Yang, Guodong Liu, Chuqin Li +10

Advanced Intelligence, Eli Lilly and Company, Indianapolis, IN, USA · Department of Biostatistics and Health Data Science, Indiana University, Indianapolis, IN, USA · Center for Biomedical Informatics, Regenstrief Institute, Indianapolis, IN, United States +2

LLM Alignment Small Language Models Preference Optimization Language Model Post-Training

Jul 13, 2026cs.CL

Direct Image-to-Modern Vietnamese Translation of Han-Nom Manuscripts via Multimodal RLHF Preference Alignment

Translating Han-Nom manuscripts into modern Vietnamese is challenging because historical pages are often degraded, the script contains rare logographic characters, and parallel supervision is limited. We propose a multimodal RLHF preference-alignment framework that conditions Vietnamese generation on manuscript images and aligned Han-Nom source text. The model combines four streams: CLIP ViT-L/14@336 for visual features, bert-base-chinese for Han-Nom representations, vinai/phobert-base for Vietnamese representations, and T5-small encoder states. Modality-specific projections and a fusion block compress the resulting 2,048-dimensional concatenation into a shared 512-dimensional representation. Starting from the same supervised fine-tuned policy, we compare PPO, DPO, and KTO under matched work-level macro-averaged evaluation. DPO achieves the best BLEU-4, ROUGE-L, BERTScore, semantic similarity, CER, WER, and token accuracy, whereas PPO obtains the highest precision, recall, and F1. KTO remains competitive through its desirable-undesirable utility objective. All preference-aligned policies improve the BLEU-4 and semantic-similarity scores available for the SFT baseline. These results indicate that multimodal preference optimization complements supervised learning by improving lexical and semantic quality in low-resource historical translation.

Thi Kim Trang Vo, Nghia Hieu Nguyen, Ha Minh Tan

University of Information Technology, Ho Chi Minh City, Vietnam · Vietnam National University, Ho Chi Minh City, Vietnam

Direct Preference Optimization Preference Optimization Machine Translation Multimodal Learning

Jul 3, 2026cs.IR

PORTS: Preference-Optimized Retrievers for Tool Selection with Large Language Models

Integrating external tools with Large Language Models (LLMs) has emerged as a promising paradigm for accomplishing complex tasks. Since LLMs still struggle to effectively manage large tool collections, researchers have begun exploring retrieval-based methods to pre-select the most relevant options, addressing input length and latency constraints. However, existing retrievers are often misaligned with tool-calling LLMs due to their separate training processes. This paper presents PORTS, a novel odds ratio preference optimization method for training retrievers aimed at tool selection. Using a perplexity-inspired preference signal from a frozen LLM, our approach fine-tunes a retriever to find helpful tools by optimizing the correlation between the selection probabilities and the downstream performances while jointly enforcing a contrastive semantic loss between documentation strings. The versatility of PORTS and its ability to significantly improve tool selection accuracy are demonstrated through extensive experiments on six datasets, two encoder models, and three LLMs with diverse prior knowledge. With low computational demands, our alignment process facilitates generalization to new queries and tools, proving valuable for practical applications with evolving toolsets.

Lorenzo Molfetta, Giacomo Frisoni, Nicolò Monaldini +1

Department of Computer Science and Engineering, University of Bologna

Preference Optimization LLM Tool Use Tool Selection

Jul 3, 2026cs.CL

KARMA: Knowledge graph-based Automated Reasoning Materialization and Alignment

Template-based contrastive synthesis is scalable, but its candidates often differ only in a few entity-slots while sequence-level optimization spreads supervision over mostly shared templates. We formalize this as the Resolution Mismatch Problem and propose KARMA, which enumerates schema-constrained paths over domain knowledge graphs and verbalizes them into slot-aligned contrastive candidates. Slot-Parallel Alignment (SPA) then applies a decoupled slot-level objective to route preference supervision to discriminative entity-slots, with slot-aware masked attention serving as an optional packed-evaluation implementation. Across biomedical, computer-science, and chemistry benchmarks, KARMA outperforms base LLM and same-data SFT baselines, and compares favorably with sequence- and token-level preference methods.

Jinkyeong Choi, Chaebin Jeong, Donghyeon Park

Sejong University

Contrastive Learning Preference Optimization Preference Alignment

Jul 2, 2026cs.AI

Distributionally Robust Listwise Preference Optimization

Existing robust preference optimization for language-model alignment mainly studies pairwise supervision and places robustness at the dataset, prompt, or preference-pair level. We instead study listwise preference optimization under ranking-label uncertainty: given a prompt and a candidate list, the observed ranking over that list may be ambiguous due to annotator inconsistency, near-ties, lossy rankwise feedback, or reward-model noise. We propose a pointwise total-variation robust Plackett--Luce objective that directly robustifies the ranking label conditional on the candidate list. The robust loss admits an exact decomposition into the nominal PL loss plus a worst-case PL correction, and the worst-case ranking is obtained by sorting current implicit scores in ascending order, reducing the inner maximization from

K!

enumeration to

O(K\log K)

. This tractable structure yields strong offline and online optimization guarantees. In the offline fixed-list setting, the robust objective is convex and projected stochastic subgradient reaches global

ε

-suboptimality with

O(ε^{-2})

sample complexity. In the online policy-induced setting, where candidate lists are generated by the current policy, we establish weak convexity and

\widetilde O(ε^{-2})

Moreau-envelope stationarity. Experiments in offline LLM alignment show that the proposed robust correction largely preserves performance under clean labels and improves robustness under noise. In online alignment, it makes reward-model-ranked candidate expansion more reliable and improves both reward-model and external GPT-4 judge metrics.

Xudong Wu, Jian Qian, Pangpang Liu +2

The University of Hong Kong · Hong Kong SAR · Yale University +3

LLM Alignment Distributionally Robust Optimization Robust Optimization Robust Loss Functions Preference Optimization

Jun 26, 2026cs.CV

RSICCLLM: A Multimodal Large Language Model for Remote Sensing Image Change Captioning

Remote Sensing Image Change Captioning (RSICC) aims to describe changes between bi-temporal remote sensing images and holds significant research and application value. However, most existing methods rely on conventional deep learning architectures, and the limited model capacity constrains performance. Although large-model post-training techniques have achieved great success in general domains, their direct transfer to RSICC remains challenging due to data scarcity and the need for fine-grained change understanding. To address this, we propose RSICCLLM, the first post-training framework for large vision-language models in RSICC. Specifically, we design a data generation paradigm, release the instruction dataset RSICI, and establish a task-specific RSICC benchmark. We further introduce Difference-aware Supervised Fine-tuning to explicitly extract change representations and guide the model in perceiving and understanding temporal differences. In addition, we propose Dual-Negative Preference Optimization (DNPO), which employs two complementary negative-sample construction strategies to construct the preference dataset RSICP and further refine model performance. Extensive experiments validate the superior capability of RSICCLLM, which achieves outstanding results with only 7B parameters, surpassing models of substantially larger scales. The code and dataset will be made publicly available at https://github.com/keaill/RSICCLLM.

Yelin Wang, Zijia Song, Shuo Ye +6

State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China · Harbin Institute of Technology, Shenzhen, China · Great Bay University, Dongguan, China +1

Remote Sensing Image Understanding VLM Adaptation Remote Sensing Image Change Captioning Multimodal Large Language Models Preference Optimization Remote Sensing Benchmarks

Jun 25, 2026cs.AI

MORPH: Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization

Embedding-based retrieval typically returns highest-scoring items, but many production scenarios require items that satisfy a target attribute while preserving a fine-grained pattern expressed by seed examples. We formalize this as pattern-preserving attribute retrieval. Standard approaches fail: averaging seeds preserves the pattern but misses the attribute; global attribute retrieval drifts to unrelated patterns. We approach the task with continuous generative retrieval, where a model reads item-embedding sequences and generates query embeddings for nearest-neighbor search. We propose MORPH: Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization, a staged framework with large-scale raw-sequence pretraining, Metric-Ordered Sequence (MOS) training, and final HPPO alignment. MOS construction turns sparse online metric labels into in-pattern trajectories; MOS CPT/SFT then trains the generator through shared multi-domain continuation pretraining and domain-specific tail-centroid supervised fine-tuning. HPPO uses a hybrid pool of static and policy-generated candidate embeddings, labels them with true online intersection metrics, applies iterated preference optimization, and employs a Pareto pair filter to exclude winners that lower pattern purity. Across four large-scale attribute domains under strict item- and pattern-holdout protocols, MOS training improves the primary intersection metric over a strong pretrained generative retriever in every domain-split cell, and the complete Pareto-filtered HPPO procedure improves it further, with paired-bootstrap-significant gains on seven of the eight cells - the exception being the D4 pattern-holdout split. Ablations confirm that the Pareto pair filter improves the attribute-pattern tradeoff on D1-D3, and that hybrid static/policy candidates are complementary.

Chenghao Liu, Yu Zhang, Zhongtao Jiang +8

Peking University

Generative Retrieval Preference Optimization

Jun 24, 2026cs.CV

OracleAnalyser: Analysing Implicit Semantics of Oracle Bone Scripts through MLLMs with Post-training

With the advancement of artificial intelligence, research on oracle bone scripts has entered a new era. However, existing methods and benchmarks remain largely confined to recognition tasks, overlooking the equally crucial aspect of oracle bone analysis. To address this gap, we propose OracleAnalyser, a reasoning framework for oracle bone analysis based on post-training techniques. Specifically, we fine-tune Qwen2.5-VL-3B-Instruct through multiple post-training stages and introduce a new preference optimization algorithm, Stable Focal Preference Optimization (SFPO), tailored to the characteristics of oracle bone datasets. In addition, we release both an oracle bone reasoning dataset and an oracle bone preference dataset, and further construct a new benchmark to evaluate models' analytical capabilities for oracle bone scripts. Extensive experiments validate the superior analytical performance of OracleAnalyser, which achieves remarkable results with only 3B parameters, surpassing models with substantially larger scales.

Zijia Song, Yelin Wang, Zhengyi Ma +5

National University of Defense Technology · ShanghaiTech University · Great Bay University +1

VLM Reasoning Preference Optimization Language Model Post-Training

Jun 22, 2026cs.AI

Against Proxy Optimization

I discuss conditions under which maximizing a proxy utility function is harmful and suggest this poses problems for applying decision theory.

Sven Neth

University of Pittsburgh

Preference Optimization Decision-Making under Uncertainty

Jun 18, 2026cs.CL

Beyond Uniform Forgetting: A Study of Sequential Direct Preference Optimization Across Preference Settings

Aligning language models with human preferences often requires optimising multiple behavioural objectives. A practical approach is to apply these objectives sequentially using preference optimisation methods such as Direct Preference Optimisation (DPO), but it remains unclear whether later training uniformly degrades preferences learned earlier or whether the effect depends on the relationship between objectives. We study sequential DPO across four preference settings covering distributional conflict, multi-attribute interaction, strong safety signal, and compatible response-quality objectives. Using Llama-3.1-8B-Instruct with LoRA adapters, we evaluate all objectives after every stage with a fixed base-model reference. We find that sequential DPO does not produce a single forgetting pattern; preference change ranges from partial degradation to stability, pair-level redistribution, or positive transfer depending on objective relationship, signal strength, and training order. Pair-level analysis using length-normalised policy margins shows that aggregate metrics can mask heterogeneous changes across preference pairs, whereas quartile decomposition reveals that high-confidence pairs can either degrade or improve depending on the setting. Mechanistic diagnostics show that Stage~2 gradients and adapter updates are near-orthogonal to the previous objective across all settings, providing little evidence that direct gradient opposition is the primary driver. These findings suggest that future sequential alignment pipelines should account for objective compatibility and signal strength, rather than assuming that later objectives affect earlier preferences uniformly.

Pranav Bhandari, Nicolas Fay, Amitava Datta +2

1Network Analysis and Social Influence Modelling (NASIM) Lab · School of Physics Maths and Computing, The University of Western Australia · School of Psychological Science, The University of Western Australia +1

Continual Learning for LLMs Direct Preference Optimization Preference Optimization Multi-Objective Language Model Alignment Catastrophic Forgetting

Jun 16, 2026cs.IR

Temporal Preference Optimization for Unsupervised Retrieval

Unsupervised dense retrievers offer scalability by learning semantic similarity from unlabeled documents via contrastive learning, but they struggle to capture the temporal relevance, retrieving semantically related but temporally misaligned documents-an important aspect when a document collection spans multiple time periods (e.g., retrieving documents from 2018-2025 for "Who is the president in 2019?" introduces temporal ambiguity). Existing methods rely on supervised training with explicit timestamps, which are not always feasible. We propose TPOUR (Temporal Preference Optimization for Unsupervised Retriever), which uses our novel training method Temporal Retrieval Preference Optimization (TRPO). TRPO reinterprets preference learning in the temporal dimension, guiding the retriever to favor temporally aligned documents. TPOUR further generalizes to unseen time periods via interpolation in a learned time embedding, enabling continuous temporal alignment. Experiments on temporal information retrieval (T-IR), TPOUR outperforms both unsupervised and supervised baselines. Compared to Qwen-Embedding-8B, despite being about 72.7x smaller, TPOUR Contriever improves average nDCG@5 by +4.04 (+12.15%) on explicit and +4.98 (+15.21%) on implicit queries. We provide our code at https://github.com/agwaBom/TPOUR.

HyunJin Kim, Jaejun Shim, Young Jin Kim +1

1Sungkyunkwan University, Suwon, South Korea · 2Microsoft, Redmond, USA

Temporal IR Preference Optimization Dense Retrieval

Jun 15, 2026cs.NE

Effects of Objective Normalization on Regions of Interest in Preference-Based Evolutionary Multi-Objective Optimization

Preference-based evolutionary multi-objective optimization (PBEMO) aims to approximate a region of interest (ROI) defined by the preference information from a decision maker (DM). Although objective functions in real-world applications typically have different scales, the issue of how to define the ROI in such problems has been overlooked in the literature. In fact, it has not been standardized in the EMO community whether the ROI should be defined in the unnormalized objective space or in the normalized objective space. In this context, this paper investigates the effects of objective normalization on ROIs. First, this paper shows that two ROIs defined in the unnormalized and normalized objective spaces can differ significantly for problems with differently scaled objectives. Then, we demonstrate that ROIs defined in the normalized objective space are highly difficult to approximate even on problems with equally scaled objectives because of poor approximations of the ideal and nadir points. In contrast, we show that ROIs defined in the unnormalized objective space are much easier to approximate than those defined in the normalized objective space.

Ryuichi Mogami, Ryoji Tanabe

Yokohama National University, Yokohama, Japan

Multi-Objective Evolutionary Optimization Preference Optimization Multi-Objective Optimization

Jun 15, 2026cs.CL

PVminerLLM2: Improving Structured Extraction of Patient Voice via Preference Optimization

Motivation: Patient-generated text contains critical information on patients' lived experiences, social context, and care engagement, but remains largely unstructured, limiting its use in patient-centered outcomes research. Prior work introduced the PV-Miner benchmark and PVMinerLLM models for structured extraction. However, supervised fine-tuning (SFT) alone struggles with rare, fine-grained, and unevenly distributed errors, particularly in token-critical structured outputs. Results: We present PVminerLLM2, an improved set of LLMs for structured patient voice extraction that applies preference optimization to address token-critical errors beyond the reach of supervised fine-tuning. Our method introduces (i) a preference objective with token-level gated stabilization term that prevents degradation of absolute token likelihood under preference optimization, and (ii) confusion-aware preference pair construction to better capture low-separation distinctions. We further incorporate token-importance weighting and inverse-frequency reweighing to address token imbalance and class skew. Across multiple model sizes, PVMinerLLM2 consistently outperforms strong baselines, achieving gains of up to 4.43% (Code), 3.50% (Sub-code), and 1.55% (Span), and outperforms baseline LLM trained with existing preference optimization methods. Availability and Implementation: The supplementary material, code, evaluation scripts, and trained models for PVminerLLM2 are publicly available at: https://github.com/Data-Mining-Lab-Yale/PVminerLLM2

Samah Fodeh, Linhai Ma, Ganesh Puthiaraju +7

Department of Emergency Medicine, Yale School of Medicine, 464 Congress Ave, 06519, CT, USA · Department of Social and Behavioral Sciences, Yale School of Public Health, 60 College Street, 06520, CT, USA · Division of Research, Texas State University, 601 University Dr., 78666, TX, USA

Healthcare Clinical Information Extraction Preference Optimization LLM Information Extraction Medical LLMs

Jun 11, 2026cs.CL

PolyAlign: Conditional Human-Distribution Alignment

Post-training methods such as supervised fine-tuning (SFT) and preference optimization typically align language models toward a single global assistant behavior. While effective for improving average helpfulness, this can suppress the natural variation of human responses across languages, tasks, and dialogue settings. We study this problem as conditional human-distribution alignment: models should match the human response distribution appropriate to the current interaction context, rather than a universal response style. We introduce PolyAlign, a distribution-aware alignment framework that organizes bilingual interaction data into bucket-specific human reference distributions defined by language, interaction track, response family, and length. PolyAlign combines Bucket-Aware SFT, which balances optimization across heterogeneous buckets, with Human-Distribution Preference Optimization (HDPO), which regularizes preference learning using critic-estimated distance to bucket-specific human support. Across a bilingual evaluation suite covering English and Chinese single- and multi-turn settings, PolyAlign improves conditional naturalness and distributional faithfulness while preserving competitive task utility. The results suggest that post-training should move beyond global alignment objectives toward interaction-aware alignment with human response distributions.

L. D. M. S. Sai Teja, Ufaq Khan, Sathira Silva +2

NIT Silchar, India · MBZUAI, Abu Dhabi, UAE

Multilingual Language Models LLM Alignment Preference Optimization Language Model Post-Training

Jun 11, 2026cs.SD

Emo-LiPO: Listwise Preference Optimization for Fine-Grained Emotion Intensity Control in LLM-based Text-to-Speech

Large language model (LLM)-based text-to-speech (TTS) systems enable prompt-conditioned emotional control but struggle with fine-grained emotion intensity due to the semantic -- acoustic gap between text and speech. To address this challenge, we formulate emotion intensity control in LLM-based TTS as a learning-to-rank problem and propose Emo-LiPO, a listwise preference optimization framework that aligns prompt-conditioned speech generation with relative emotion intensity expressed in text. Emo-LiPO explicitly models global intensity ordering within each emotion under fixed transcripts, enabling more faithful and continuous emotional expression. We further construct ESD-plus, a multi-speaker dataset with explicit emotion intensity variations, to support fine-grained emotion modeling and evaluation. Experiments on ESD-plus demonstrate that Emo-LiPO significantly improves emotion accuracy and intensity controllability over both supervised- and DPO-based LLM TTS baselines, with particularly pronounced gains at high intensity levels.

Yihang Lin, Li Zhou, Congwei Cao +4

The Chinese University of Hong Kong, Shenzhen · Shenzhen Loop Area Institute · Agency for Science, Technology and Research +2

TTS Synthesis Preference Optimization Controllable Speech Generation Emotional Speech Synthesis

Jun 9, 2026cs.CL

SAGE: Answer-Conditioned Uncertainty Targets for Verbal Uncertainty Alignment

Large language models increasingly express uncertainty through natural-language statements, yet these expressions often fail to reflect the model's sampled behavior. We study verbal uncertainty alignment as a distributional calibration problem: the appropriate uncertainty target for a prompt should be estimated from repeated model outputs rather than from an isolated response. However, group rollouts alone are insufficient, since the resulting target must provide a useful training signal. Existing targets only partially satisfy this requirement. We propose SAGE, Semantic-Answer Guided Entropy, a group-level uncertainty target that constructs an answer-conditioned uncertainty geometry over sampled responses. SAGE preserves categorical, numeric, and symbolic answer distinctions while maintaining a smooth and scale-preserving calibration signal. We further apply this target through Group-Uncertainty Preference Optimization, or GUPO, an uncertainty-channel training framework that supervises verbal uncertainty expressions rather than the full response. Experiments across factual, mathematical, and multiple-choice reasoning tasks show improved uncertainty ranking, lower calibration error, and reduced overconfidence.

Kaiwen Shi, Zheyuan Zhang, Yanfang Ye

University of Notre Dame

Confidence Estimation in Language Models Preference Optimization LLM Uncertainty Estimation Uncertainty Calibration

Jun 9, 2026cs.CV

FoA-SR: Faithful or Aesthetic? Profile-Aware Preference Optimization for Real-World Image Super-Resolution

Real-world image super-resolution (SR) is often designed with a single restoration objective, despite the current capacity of generative models to produce multiple high-quality reconstructions for the same input. In this paper, we argue that the best restoration strategy is subject to the specific restoration profile: a Faithful restoration prioritizes reference consistency, structure preservation, and hallucination suppression, whereas an Aesthetic restoration prioritizes visually pleasing and natural-looking details. We propose FoA-SR, a novel preference optimization approach to real-world SR based on profiles. To achieve this goal, FoA-SR starts with our supervised FLUX.2-based SR adapter (Flux2SR) trained with LR latent conditioning, flow matching, and image-space reconstruction losses for paired LR-to-HR image super-resolution. Following the development of the shared supervised super-resolution adapter, FoA-SR generates a shared stochastic candidate pool for each input image and ranks the same candidates using profile-specific Faithful and Aesthetic rewards to mine winner-loser pairs. These pairs are used to fine-tune separate LoRA adapters while keeping the base model frozen. Experiments on RealSR and DIV2K show that FoA-SR can steer the same SR adapter towards distinct restoration objectives: a Faithful adapter improves reference-consistent metrics while an Aesthetic adapter boosts metrics that measure perceptual quality without reference. Our candidate-pool analysis shows that Faithful and Aesthetic rewards frequently select different winners, and a Hybrid-LoRA ablation shows that collapsing both profiles into one reward yields an implicit compromise rather than explicit profile control.

Amjad Mahdi Alqarni, Peizhong Ju

Department of Computer Science University of Kentucky Lexington, KY, USA

Diffusion Models for Image Restoration Preference Optimization Image Super-Resolution

Jun 2, 2026cs.AI

ThoughtFold: Folding Reasoning Chains via Introspective Preference Learning

Large Reasoning Models (LRMs) have achieved remarkable progress thanks to Reinforcement Learning with Verifiable Rewards (RLVR) on Chain-of-Thoughts (CoTs). However, since long CoTs naturally contain trial and errors and mainstream RLVR approaches choose outcome-correct CoT trajectories for memorization, the redundant explorations in long CoTs are inevitably reinforced, which results in the over-thinking issues of LRMs. Previous attempts to resolve this issue mainly give more advantage to shorter trajectories, yet their learning signals are still outcome-based and cannot reduce the memorization of redundant explorations in long CoTs. Therefore, we propose ThoughtFold, a framework that leverages fine-grained preference learning to mitigate redundant explorations for efficient reasoning. ThoughtFold employs an introspective strategy to identify redundancy within each correct trajectory, which yields a spectrum of candidate sub-trajectories. Leveraging this spectrum, we introduce a masked preference optimization objective that explicitly penalizes redundant explorations and encourages the model to directly bridge essential reasoning segments, effectively folding its reasoning chains into a more concise path. Extensive experiments show that ThoughtFold significantly enhances efficiency. It reduces the token usage of DeepSeek-R1-Distill-Qwen-7B by approximately 56% while maintaining state-of-the-art accuracy.

Ziyan Liu, Xueda Shen, Yuzhe Gu +7

Shanghai Artificial Intelligence Laboratory · University of Science and Technology of China · MMLab, The Chinese University of Hong Kong

Efficient Language Model Reasoning Preference Optimization Efficient Language Model Inference CoT Compression LLM Reasoning Large Reasoning Models

Jun 1, 2026cs.LG

Drifting Preference Optimization for One-Step Generative Models

One-step text-to-image generators are attractive for deployment because they generate an image with a single forward pass, but preference finetuning them remains difficult: standard alignment methods often rely on policy likelihoods, denoising trajectories, differentiable reward gradients, or test-time optimization. We propose Drifting Preference Optimization (DrPO), an online preference-finetuning method for deterministic one-step generators. For each prompt, DrPO samples candidates from the current generator, ranks them with a target reward, and uses high- and low-scoring samples to synthesize a feature-space update direction. The update is a non-parametric dipole preference field plus a reference drift estimated from the frozen base generator, and is optimized through a detached feature-space regression target. The target reward is used only for ranking, so DrPO can train with large, black-box, or non-differentiable rewards while inference remains a single generator call. We evaluate DrPO on SD-Turbo and SDXL-Turbo with multiple target rewards and benchmarks, including HPSv3 and GenEval. DrPO improves alignment over reward-gradient-free one-step preference baselines and reduces HPSv3 training computation by

3.51\times

under the matched effective-batch setting by removing reward-model backpropagation. Initial offline experiments suggest that sample-based gradient synthesis can also be used beyond online reward ranking.

Zhou Jiang, Yandong Wen, Zhen Liu

Westlake University · The Chinese University of Hong Kong, Shenzhen

Diffusion Model Alignment One-Step Generative Modeling Preference Optimization T2I Diffusion Models

May 29, 2026cs.AI

From "Weak" Signals to Strong Models: Preference Delta Aggregation with LoRA Merging

Training strong large language models (LLMs) requires high-quality supervision, which is often scarce. Recent work shows that paired preference data from weak-weaker model pairs (e.g., Qwen3 4B over 1.7B), despite the limited quality of individual responses, can provide an effective supervision signal through relative quality deltas, which we term a "weak" signal. This motivates a key research question: can multiple "weak" signals be constructively aggregated for improving strong models (e.g., Qwen3 8B)? To this end, we propose Preference Delta Aggregation (PDA), the first framework that derives a preference delta from each weak-weaker model pair, instantiates it as a LoRA adapter learned through preference optimization, and aggregates the resulting deltas via LoRA merging. To further mitigate directional interference during LoRA merging, we introduce Geometric Alignment Merging (GAM), a geometry-aware merging method that aligns adapter subspaces before aggregation, enabling more robust composition of diverse deltas. Evaluations on knowledge reasoning and agentic search benchmarks show that aggregating multiple "weak" signals pushes performance beyond any single signal, with further gains as additional signals are incorporated. Correspondingly, PDA with GAM improves the strong model by 6.8 and 7.3 points on average for knowledge reasoning and agentic search, respectively. It outperforms all single-delta and multi-delta baselines, exceeding the best single-delta baseline by 2.1 and 4.3 points. Further analysis attributes these gains to the effective composition of complementary capabilities encoded across distinct preference deltas.

Qi Sun, Siyue Zhang, Yulin Chen +3

NYU Shanghai · NTU · NYU +1

Pairwise Preference Learning Low-Rank Adaptation Preference Optimization Language Model Merging

May 27, 2026cs.CL

Human Label Variation as Stable Signal: Learning Annotator-Specific Explanation Behavior via Cross-Annotator Preference Optimization

Free-text explanations extend human label variation (HLV) beyond label disagreement by revealing the reasoning and preferences behind annotators' decisions. We study whether large language models (LLMs) can learn and reproduce such annotator-specific label-explanation behavior. Using two sentence-pair tasks with four annotators each -- natural language inference and paraphrase judgment -- we first analyze whether annotators exhibit stable individual patterns. We find that such patterns are weak at the single-annotation level due to strong input-content effects, but become detectable after input-content reduction and annotator-level aggregation. We then compare prompting and supervised fine-tuning (SFT) baselines and propose cross-annotator preference optimization (CAPO), which contrasts a target annotator's response with other valid but less target-specific annotations for the same input. Experiments show that prompting is limited and unstable, SFT better captures annotator-specific behavior, and CAPO further improves aggregation-aware imitation and judge-based attribution while preserving target-specific reasoning patterns under human validation. Overall, our results show that HLV can be learned as annotator-specific label-explanation behavior, suggesting a path toward scalable explanation-based annotation grounded in annotator histories rather than labels alone.

Beiduo Chen, Pingjun Hong, Ziyun Zhang +3

U MaiNLP, Center for Information and Language Processing, LMU Munich, Germany · Æ Munich Center for Machine Learning (MCML), Munich, Germany · [ University of Vienna, Austria +1

LLM-Assisted Annotation Preference Optimization Text Generation