Explore the Frontier of Global Academia

01.

arXiv (quant-ph) 2026-06-16 DOI: arXiv:2601.18872

Against probability: A quantum state is more than a list of probability distributions

Authors:

Ladina Hausmann↗Renato Renner↗

arXiv:2601.18872v2 Announce Type: replace Abstract: The state of a quantum system can be represented by listing the outcome probabilities for a tomographically complete set of measurements. Such representations appear throughout physics, for example, in quantum field theory via correlation functions and in quantum foundations within generalized probabilistic frameworks. In this paper, we show a no-go result: To enable useful statements, the probability representation must be topologically robust$\unicode{x2014}$preserving the notion of closeness between states. Yet, a topologically robust probability representation cannot simultaneously retain other essential structure, such as the subsystem structure.

Read & Discuss → View Source →

02.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2601.22108

Learning What to Predict: Downstream-Guided Task Design for Continued Pretraining

Authors:

Shuqi Ke↗Giulia Fanti↗

arXiv:2601.22108v2 Announce Type: replace-cross Abstract: Continued pretraining is optimized with fixed self-supervised tasks but selected by downstream performance, creating a coarse feedback loop in which practitioners evaluate checkpoints, change data mixtures or objectives, and restart runs, while individual updates remain blind to target capabilities. We ask whether a small set of verifiable downstream examples can provide step-level feedback without directly supervising the learner. We introduce V-pretraining, which decouples a learner trained only with a self-supervised loss from a lightweight task designer that constructs targets or views for unlabeled batches. Given the current learner and batch, V-pretraining scores a candidate construction by predicting the first-order reduction in downstream loss after the induced self-supervised update. The designer maximizes this value; the learner then applies the update with targets or views detached, so downstream labels never update learner parameters. We instantiate V-pretraining as adaptive top-K soft targets for language modeling and learned views or masks for self-supervised vision. Across both modalities, V-pretraining improves target capabilities without degrading generalization. Under wall-clock-matched continued pretraining, it improves GSM8K Pass@1 for Qwen models using 1,024 GSM8K examples only as feedback, including a +7.4 point single-run gain for Qwen2.5-0.5B. In vision, it improves DINOv3 transfer to ADE20K semantic segmentation and NYUv2 depth estimation while preserving ImageNet linear accuracy, suggesting that feedback-guided task construction can improve target capabilities without collapsing general-purpose representations.

Read & Discuss → View Source →

03.

arXiv (CS.CL) 2026-06-15 DOI: arXiv:2606.14368

Be My Tutor: On-Policy Co-Distillation for Mutual LLM Improvement via Peer Feedback

Authors:

Woohyeon Byeon↗Jiwon Jeon↗Jeonghye Kim↗Youngchul Sung↗

We study multi-domain LLM training in which two models, each stronger in a different domain, co-evolve by tutoring each other through on-policy feedback. Unlike one-way distillation or single-model fine-tuning, our goal is mutual Pareto improvement: each model improves across domains without losing its original strength. To this end, we propose On-Policy Co-Distillation (OPCoD), where each student's self-distillation is conditioned on its own correct rollout and feedback from its peer. To make feedback exchange effective, OPCoD uses cognizance-based gating to decide when to give feedback and feedback anchoring to ground feedback in the problem. On Science Q\&A tasks, OPCoD consistently outperforms baselines and achieves Pareto improvement across all evaluated domain pairs and students.

Read & Discuss → View Source →

04.

arXiv (CS.LG) 2026-06-18 DOI: arXiv:2606.19057

Quantifying and Auditing LLM Evaluation via Positive–Unlabeled Learning

Authors:

Zilong Zhang↗Yi-Ting Hung↗Lei Ding↗Chi-Kuang Yeh↗

arXiv:2606.19057v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used as judges for scalable evaluation, yet such LLM–as–a–Judge systems exhibit systematic biases that are decoupled from semantic quality, most notably verbosity bias. Meanwhile, human supervision is costly and typically selective, yielding reliable positive judgments but leaving most outputs unlabelled and potentially mixed in quality. We formulate LLM evaluation under selective human supervision as a positive–unlabelled learning problem and propose a geometric auditing framework based on Partial Optimal Transport. By aligning a small set of human–verified positives with a reliable subset of unlabelled outputs in a fixed embedding space, our method identifies human–consistent preferences and corrects biased judges without retraining. Experiments demonstrate improved alignment with human preferences, increased robustness to presentation biases, and interpretable confidence estimates, offering a scalable and statistically grounded alternative to existing LLM–as–a–judge pipelines.

Read & Discuss → View Source →

05.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2606.12294

Bridging the Modality Gap in Forensic Image Retrieval

Authors:

Ricardo Gonz\'alez-Gazapo↗Annette Morales-Gonz\'alez↗Yoanna Mart\'inez-D\'iaz↗Heydi M\'endez-V\'azquez↗Milton Garc\'ia-Borroto↗

Automated image retrieval plays an increasingly critical role in modern forensic analysis, supporting investigative workflows that rely on efficient comparison of visual evidence. While prior work has focused primarily on developing and optimizing multimodal retrieval systems, limited attention has been paid to evaluating the forensic applicability of these technologies across diverse real-world scenarios. In this study, we present a unified retrieval framework adapted to four key forensic tasks: (1) tattoo image retrieval given a tattoo query image; (2) tattoo retrieval guided by human-expert textual descriptions, modelling the common situation where a witness verbally describes a tattoo; (3) tattoo retrieval from hand-drawn sketches; and (4) face retrieval from forensic face sketches. Our system leverages a multimodal large language model (MLLM) to automatically generate structured textual descriptions for all queries and gallery images, followed by sentence-transformer embedding for text-based comparison. We evaluate retrieval using visual-only embeddings, text-only embeddings and a multimodal fusion strategy that combines text- and image-based similarity scores derived from state-of-the-art visual feature extractors relevant to each task. The fusion of modalities consistently improves retrieval precision and robustness, especially in scenarios where visual information is limited or noisy (e.g., sketches, partial tattoos, or fragmented witness statements). This work highlights the forensic value of a unified multimodal retrieval pipeline and demonstrates how modern MLLMs can operationalize challenging forensic tasks that traditionally rely on manual expert analysis. Our results position multimodal retrieval as a promising tool for supporting investigative workflows involving tattoos, facial composites, and witness descriptions.

Read & Discuss → View Source →

06.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2605.17443

Analyzing Error Propagation in Korean Spoken QA with ASR-LLM Cascades

Authors:

Donghyuk Jung↗Youngwon Choi↗

We analyze how automatic speech recognition (ASR) errors propagate through ASR-LLM cascades in Korean spoken question answering (SQA), focusing on downstream semantic failures that conventional ASR metrics cannot fully capture. Our analysis shows that the relative downstream degradation caused by ASR errors is consistent across LLMs with different absolute performance, suggesting that cascade degradation largely tracks ASR-stage information loss. We further identify single-character Korean ASR errors as a Korean-specific loss channel, where even a minimal transcription difference can change the intended question and degrade downstream QA performance. Finally, an auxiliary comparison shows that a large audio language model outperforms an ASR-LLM cascade with an approximately matched language backbone in noisy Korean SQA, indicating the potential of direct audio input to mitigate transcript-induced information loss.

Read & Discuss → View Source →

07.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2606.17598

MuseVLA: An Adaptive Multimodal Sensing Vision-Language-Action Model for Robotic Manipulation

Authors:

Xingyuming Liu↗Ruichun Ma↗Heyu Guo↗Qixiu Li↗Qingwen Yang↗Lin Luo↗Shiqi Jiang↗Chenren Xu↗Jiaolong Yang↗Baining Guo↗

Humans naturally leverage diverse sensing modalities to interact with the physical world, while most Vision-Language-Action (VLA) models for robotics rely solely on RGB observations. This limits their ability to perceive physical properties that are difficult or impossible to infer from RGB cameras, such as temperature, sound, or radar response. We present MuseVLA, an adaptive multimodal sensing VLA model that integrates novel sensors as on-demand tools for robotic manipulation. Given a task instruction and visual context, MuseVLA first generates a sensor token and target description that select the sensing modality to invoke and what to attend to, analogous to a tool call with arguments. It then converts the selected sensor measurement into a grounded sensor image, a unified intermediate representation that encodes heterogeneous readings for multimodal fusion and action generation. This design decouples sensor-specific processing from the VLA backbone, enabling efficient integration of diverse modalities. To reduce the need for expensive multisensory robot datasets, we further introduce a data synthesis pipeline that augments existing RGB video datasets with grounded sensor images, enabling generalization to unseen sensor-guided tasks. We evaluate MuseVLA on a real-world robot across challenging dexterous hand manipulation tasks that require multimodal sensing inputs, including temperature-guided pick-and-place, audio-driven object search, and radar-assisted hidden object retrieval. MuseVLA achieves 80.6% success rate on average, outperforming RGB-only and multisensory VLA baselines significantly, and exhibits strong zero-shot capabilities on unseen tasks.

Read & Discuss → View Source →

08.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.15207

Controlled Dynamics Attractor Transformer

Authors:

Cheng Zhang↗Minnan Luo↗Zesheng Yang↗Ming Li↗Yong-Jin Liu↗Qinghua Zheng↗

arXiv:2606.15207v1 Announce Type: cross Abstract: Transformer architectures have dramatically advanced representation learning and inference in deep models through self-attention mechanisms. In parallel,associative memory (AM) frameworks map representations onto energy landscapes, offering interpretable retrieval mechanisms. However, their continuous-time inference dynamics lack the biological plausibility of classical Continuous Attractor Neural Networks (CANNs). To bridge this gap, we propose Controlled Dynamics Attractor Transformer (CDAT), which couples a mixture von Mises-Fisher (Mo-vMF) attention energy with a Hopfield refinement energy, while augmenting energy descent with a CANN-inspired excitation-inhibition modulation. CDAT instantiates a topology-constrained dynamical system whose couplings encode relational structure among tokens, thereby linking attractor-style dynamics to modern energy-based attention. We further provide a constructive dissipation analysis to formally establish their controlled inference dynamics. Benefiting from these robust and structured dynamics, CDAT achieves state-of-the-art performance across multiple benchmarks in graph anomaly detection and graph classification.

Read & Discuss → View Source →

09.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.15416

Encode Errors: Representational Retrieval of In-Context Demonstrations for Multilingual Grammatical Error Correction

Authors:

Guangyue Peng↗Wei Li↗Wen Luo↗Houfeng Wang↗

Grammatical Error Correction (GEC) involves detecting and correcting the wrong usage of grammar. While large language models (LLMs) with in-context learning (ICL) capabilities have shown significant progress on various natural language processing (NLP) tasks, their few-shot performance on GEC remains suboptimal. This is mainly due to the challenge of retrieving suitable in-context demonstrations that capture error patterns instead of semantic similarity. In this paper, we demonstrate that LLMs can inherently capture information related to grammatical errors through their internal states. From these states, we extract the Grammatical Error Representation (GER), an informative and semantically neutral encoding of grammatical errors. Our novel GER-based retrieval method significantly boosts performance in ICL settings on multilingual GEC datasets, improving the precision of correction. For high-resource languages, our results on 8B-sized open-source models match those of closed-source models such as Deepseek2.5 and GPT-4o-mini. For low-resource languages, our $F_{0.5}$ scores surpass the baseline by up to a factor of 1.20. This method provides a more precise and resource-efficient solution for multilingual GEC, offering a promising direction for interpretable GEC research.

Read & Discuss → View Source →

10.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.19749

Benchmarking Agentic Review Systems

Authors:

Dang Nguyen↗Wanqing Hao↗Yanai Elazar↗Chenhao Tan↗

A new class of agentic review systems are emerging as a remedy to the pressure placed on peer review systems by AI-assisted research, but it is unclear how they should be evaluated. We evaluate two open-source systems (OpenAIReview and coarse), one proprietary system (Reviewer3), and a zero-shot baseline, across six LLMs spanning frontier and efficient models. First, we study whether AI reviews on ICLR/NeurIPS papers track with papers' quality as approximated by external signals such as citations and acceptance decisions. Every system performs above chance in pairwise accuracy, and the best is OpenAIReview + GPT-5.5 at 83.0%. Second, to test whether systems can catch errors with known ground truth, we construct a perturbation benchmark that injects four categories of errors into papers across eight arXiv subject classes and measure detection recall. The strongest configuration (OpenAIReview + GPT-5.5) catches 71.6% of injected errors, leaving substantial room for improvement. The union of detections across six models reaches 83.3% recall, suggesting different models detect different errors and better harness design can potentially increase performance. Beyond these benchmarks, we study a public deployment of OpenAIReview with real users. Votes on its comments skew positive at 1.44 to 1, and the most common complaints are about false positives and minor nitpicks. Together, by evaluating full review systems backed by state-of-the-art models on real research papers, we show that while AI reviews still have room for improvement, they can already track human quality judgments well, catch important errors, and earn positive feedback from real users.

Read & Discuss → View Source →

11.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.19259

A Multi-Domain Benchmark for Detecting AI-Generated Text-Rich Images from GPT-Image-2

Authors:

Yijin Wang↗Shuyi Wang↗Wenhan Zhang↗Yuqi Ouyang↗

Text-rich images often contain privacy-sensitive, transactional, or decision-relevant information. As recent multimodal image generation models become increasingly capable of synthesizing realistic textual content and structured visual designs, detecting AI-generated text-rich images has become an important challenge for digital trust and content authenticity. Existing benchmarks, however, largely focus on object-centric images and provide limited coverage of scenarios where textual semantics and layout organization are central. In this paper, we introduce a multi-domain benchmark for detecting text-rich images generated by OpenAI's GPT Image 2. The benchmark contains 8,602 images across six representative categories: commercial posters, infographics, academic posters, receipts, tables, and UI screenshots. Using this benchmark, we evaluate five representative AI-generated image detectors in a zero-shot setting and analyze their overall, category-wise, and post-processing robustness. Our results show that detector performance is highly domain-dependent: methods that perform well in some categories often fail on others, and even the strongest conventional detector exhibits severe sensitivity to JPEG compression. We further conduct an exploratory evaluation with a multimodal vision-language model, revealing both its promise and its limitations on structured formats. These findings highlight the need for text- and layout-aware detection methods for modern AI-generated images. Our dataset is released at XXX.

Read & Discuss → View Source →

12.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2606.13817

FlowMo-WM: A World Model with Object Momentum and Hidden Ambient Drift

Authors:

Yitao Jiang↗Luyang Zhao↗Muhao Chen↗Devin Balkcom↗

arXiv:2606.13817v1 Announce Type: cross Abstract: World models in robot learning predict future states from visual observations and actions, enabling agents to reason about the consequences of their controls. However, many action-conditioned models are evaluated in settings where motion is dominated by immediate control, whereas aquatic surface vehicles and other real-world objects continue moving under inertia and are displaced by hidden ambient drift, such as water currents or wind. We propose FlowMo-WM, an end-to-end trainable visual world model that infers object-centric motion state and a predictive long-history context associated with hidden drift from image-action histories without direct supervision of flow fields. FlowMo-WM factorizes image-action history into a short-history latent state, trained to summarize object-centric motion, and a longer-history context, trained to summarize slowly varying exogenous influences. A zero-context residual transition separates action-conditioned base dynamics from context-dependent drift effects during latent rollout. In simulated aquatic surface-vehicle environments with diverse hidden flows, disturbances, and randomized vehicle dynamics, FlowMo-WM improves long-horizon rollout accuracy over representative action-conditioned latent world models. Prediction-time context ablations, in which the inferred context is zeroed or shuffled during rollout, show that the ambient context is important for stable prediction under hidden drift, while frozen linear probes characterize information encoded in the learned factors.

Read & Discuss → View Source →

13.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.20457

Repurposing a Speech Classifier for Guided Diffusion-Based Speech Generation

Authors:

Rostislav Makarov↗Timo Gerkmann↗

arXiv:2606.20457v1 Announce Type: cross Abstract: Classifier guidance is a way to control diffusion generation by using a noise-conditioned classifier to steer the sampling process toward a target class. One drawback of classifier guidance is that it requires two separately trained models: a classifier and a diffusion model. We therefore study a more compact alternative in which a conventionally trained speech classifier is repurposed as the backbone for diffusion generation. Starting from a frozen noise-conditioned classifier in log-Mel space, we attach a lightweight subnetwork that reuses intermediate classifier representations and train only this subnetwork under a Denoising Score Matching objective. Our work shows that a pretrained classifier can be repurposed for conditional generation, providing an appealing bridge between discriminative modeling and conditional speech synthesis resulting in high speech quality within a single-backbone model, with reduced memory footprint and computational cost.

Read & Discuss → View Source →

14.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.17040

R2RDreamer: 3D-aware Data Augmentation for Spatially-generalized 2D Manipulation Policies

Authors:

Xiuwei Xu↗Haowen Sun↗Angyuan Ma↗Yiwei Zhang↗Zhenyu Wu↗Xiaofeng Wang↗Bingyao Yu↗Zheng Zhu↗Jie Zhou↗Jiwen Lu↗

Spatial generalization is critical for imitation-learned manipulation policies, but achieving it typically requires scaling demonstrations across diverse object poses, robot configurations, and camera viewpoints. Data augmentation from a few source demonstrations offers a practical alternative to costly real-world collection. Simulation-based augmentation can create controllable variation, but requires complex environment and object setup and may introduce a sim-to-real gap. Recent real-to-real methods avoid these issues by jointly editing 3D observations and action trajectories from real demonstrations, yet they still rely on strong 3D scene parsing and geometry completion, and often produce observations tailored to 3D pointcloud policies rather than RGB-based 2D policies. We propose R2RDreamer, a real-to-real demonstration augmentation framework that preserves the geometric consistency of 3D action-observation editing while moving visual completion to 2D video space. Specifically, R2RDreamer first performs lightweight 3D augmentation by editing incomplete object pointclouds and end-effector trajectories in a shared 3D frame; it then projects the edited scene into masked image-space control videos with occlusion-aware reasoning and uses a dense-control image-to-video model to complete temporally coherent RGB observations. Experiments on spatially shifted manipulation tasks with both 2D diffusion-style policies and vision-language-action policies show that R2RDreamer improves spatial generalization from limited source demonstrations, with analyses validating the contributions of 3D editing, occlusion-aware projection, and video completion.

Read & Discuss → View Source →

15.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2604.09679

HCP-MAD:Heterogeneous Consensus-Progressive Reasoning for Efficient Multi-Agent Debate

Authors:

Yiqing Liu↗Hantao Yao↗Wu Liu↗Allen He↗Yongdong Zhang↗

arXiv:2604.09679v2 Announce Type: replace-cross Abstract: Multi-Agent Debate (MAD) is a collaborative framework in which multiple agents iteratively refine solutions through the generation of reasoning and alternating critique cycles. Current work primarily optimizes intra-round topologies and inter-round interactions separately, limiting the adaptation of token costs to task complexity. This work introduces Heterogeneous Consensus-Progressive Reasoning for Efficient Multi-Agent Debate (HCP-MAD), leveraging consensus as a dynamic signal to facilitate progressive reasoning. The core motivation is that a majority of straightforward tasks can be effectively resolved via lightweight pair-agent debates, while complex tasks require expanded collaboration. Firstly, Heterogeneous Consensus Verification conducts rapid consensus verification using a pair of heterogeneous agents for early stopping. Next, Heterogeneous Pair-Agent Debate applies an adaptive stopping criterion to terminate mutual critique of reasoning traces. Finally, the unresolved tasks are addressed through Escalated Collective Voting by aggregating diverse perspectives from additional agents. Experiments across six benchmarks show that HCP-MAD enhances accuracy while substantially reducing token costs. Code is https://github.com/fuyu66/HCP-MAD.

Read & Discuss → View Source →

16.

arXiv (CS.LG) 2026-06-17 DOI: arXiv:2604.01197

Learning and Generating Mixed States Prepared by Shallow Channel Circuits

Authors:

Fangjun Hu↗Christian Kokail↗Milan Kornja\v{c}a↗Pedro L. S. Lopes↗Weiyuan Gong↗Sheng-Tao Wang↗Xun Gao↗Stefan Ostermann↗

arXiv:2604.01197v4 Announce Type: replace-cross Abstract: Learning quantum states from measurement data is a central problem in quantum information and computational complexity. In this work, we study the problem of learning to generate mixed states on a finite-dimensional lattice. Motivated by recent developments in mixed state phases of matter, we focus on arbitrary states in the trivial phase. A state belongs to the trivial phase if there exists a shallow preparation channel circuit under which local reversibility is preserved throughout the preparation. We prove that any mixed state in this class can be efficiently learned from measurement access alone. Specifically, given copies of an unknown trivial phase mixed state, our algorithm outputs a shallow local channel circuit that approximately generates this state in trace distance. The sample complexity and runtime are polynomial (or quasi-polynomial) in the number of qubits, assuming constant (or polylogarithmic) circuit depth and gate locality. Importantly, the learner is not given the original preparation circuit and relies only on its existence. Our results provide a structural foundation for quantum generative models based on shallow channel circuits. In the classical limit, our framework also inspires an efficient algorithm for classical diffusion models using only a polynomial overhead of training and generation.

Read & Discuss → View Source →

17.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16183

LLM-Powered Virtual Population for Demand Simulation and Pricing

Authors:

Chengpiao Huang↗Kaizheng Wang↗

We develop an LLM-powered virtual population model that simulates demand for pricing decisions, in settings where products are described by rich unstructured information, such as text descriptions and images, and where decision makers need not only mean-demand predictions but also uncertainty estimates for counterfactual prices. Our model represents exposed customers as draws from a finite mixture of customer personas. For each persona, product, and candidate price, an LLM elicits a persona-level purchase probability using both structured persona information and unstructured product information. These probabilities are aggregated through calibrated mixture weights to form a predictive distribution of aggregate demand. The resulting simulator can evaluate counterfactual prices under various pricing objectives, including expected revenue and risk-aware criteria such as conditional value at risk. We test the framework on an online H&M fashion dataset with product descriptions and images. The calibrated LLM-based simulator achieves the best overall predictive performance among the models considered, and supports sample-efficient pricing decisions. Our framework provides a practical way to use LLMs as demand simulators for products with limited historical demand data but rich product information. By producing a full predictive demand distribution rather than only a point forecast, it enables managers to compare candidate prices, quantify demand uncertainty, and choose prices that target either average-case revenue or risk-aware objectives.

Read & Discuss → View Source →

18.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2512.13069

Multi-fidelity aerodynamic data fusion by autoencoder transfer learning

Authors:

Javier Nieto-Centenero↗Esther Andr\'es↗Rodrigo Castellanos↗

arXiv:2512.13069v2 Announce Type: replace Abstract: Accurate aerodynamic prediction often relies on high-fidelity simulations; however, their prohibitive computational costs severely limit their applicability in data-driven modeling. This limitation motivates the development of multi-fidelity strategies that leverage inexpensive low-fidelity information without compromising accuracy. Addressing this challenge, this work presents a multi-fidelity deep learning framework that combines autoencoder-based transfer learning with a newly developed Multi-Split Conformal Prediction (MSCP) strategy to achieve uncertainty-aware aerodynamic data fusion under extreme data scarcity. The methodology leverages abundant Low-Fidelity (LF) data to learn a compact latent physics representation, which acts as a frozen knowledge base for a decoder that is subsequently fine-tuned using scarce HF samples. Tested on surface-pressure distributions for NACA airfoils (2D) and a transonic wing (3D) databases, the model successfully corrects LF deviations and achieves high-accuracy pressure predictions using minimal HF training data. Furthermore, the MSCP framework produces robust, actionable uncertainty bands with pointwise coverage exceeding 95%. By combining extreme data efficiency with uncertainty quantification, this work offers a scalable and reliable solution for aerodynamic regression in data-scarce environments.

Read & Discuss → View Source →

19.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.16771

GD$^2$PO: Mitigating Multi-Reward Conflicts via Group-Dynamic reward-Decoupled Policy Optimization

Authors:

Haotian Liu↗Yihao Liu↗Jingwei Ni↗Siyuan Huang↗Xinpeng Liu↗Pengyu Cheng↗Jiajun Song↗Ruijin Ding↗Junfeng Li↗Zhechao Yu↗Mengyu Zhou↗Hongteng Xu↗…

arXiv:2606.16771v1 Announce Type: new Abstract: As LLMs advance, post-training reinforcement learning (RL) increasingly relies on multi-dimensional rewards to cultivate comprehensive capabilities. This shift demands new algorithms capable of optimizing diverse and potentially competing objectives simultaneously. To address this, existing methods such as Group reward-Decoupled Policy Optimization (GDPO) decompose the overall score into independent reward groups, then compute the RL loss separately within each group. However, this strategy still encounters multi-reward conflicts: a single rollout can yield positive advantages on certain reward dimensions but negative ones on others, causing opposing signals to cancel each other out during aggregation, further hindering RL training efficiency. Inspired by Dynamic sAmpling Policy Optimization (DAPO), which improves RL training efficiency by filtering out ineffective rollouts with near-zero advantages, we propose Group-Dynamic reward-Decoupled Policy Optimization (GD$^2$PO). Specifically, GD$^2$PO employs a conflict-aware filtering mechanism to mask out rollouts suffering from severe reward-wise disagreement. By preventing conflicting signals from canceling each other out, this masking strategy preserves and enhances the magnitude of effective RL advantages, thereby significantly accelerating learning efficiency. Furthermore, we introduce query-level reweighting to dynamically adjust the update intensity of each query based on its overall reward consensus. Experiments on various multi-reward scenarios, including tool calling and human preference alignment, demonstrate that GD$^2$PO consistently and significantly outperforms existing baselines. The code is available at https://github.com/Qwen-Applications/GD2PO.

Read & Discuss → View Source →

20.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.11915

Quality Adaptive Angular Margin Learning for Respiratory Sound Classification

Authors:

Yoon Tae Kim↗Heejoon Koo↗Miika Toikkanen↗June-Woo Kim↗

arXiv:2606.11915v1 Announce Type: cross Abstract: We present a quality-adaptive angular-margin learning framework that improves feature generalization by enforcing intra-class compactness and inter-class separability. Our framework, titled QLung, introduces a no-reference audio quality margin derived from spectral entropy and root-mean-square energy, which adaptively scales angular margins based on recording quality. To this end, we propose a log-scaled angular margin that stabilizes training under severe class imbalance. We also use an angular classifier that normalizes features and class weights, ensuring margin penalties are applied consistently on the unit hypersphere. Our approach improves in-distribution performance on the ICBHI dataset by 2.46\% over the cross-entropy baseline, and most significantly, achieves the strongest out-of-distribution performance on the SPRSound dataset compared to prior state-of-the-art methods. Code is available at https://github.com/RSC-Toolkit/QLung.

Read & Discuss → View Source →

21.

arXiv (quant-ph) 2026-06-12 DOI: arXiv:2602.10896

Toward quantum-noise-limited interferometric measurements of optical nonlinearity in vacuum

Authors:

Ali Aras↗Adrien E. Kraych↗Xavier Sarazin↗Elsa Baynard↗Fran\c{c}ois Couchot↗Moana Pittman↗

arXiv:2602.10896v2 Announce Type: replace-cross Abstract: Quantum Electrodynamics predicts that the vacuum must behave as a nonlinear optical medium: the vacuum optical index should increase when it is stressed by intense electromagnetic fields. The DeLLight (Deflection of Light by Light) project aims to measure it by using intense and ultra-short laser pulses. The experiment uses a Sagnac interferometer to amplify the tiny deflection signal of a low-intensity probe pulse crossing the vacuum refractive-index gradient produced by an external high-intensity pump pulse. The measurement of the amplified signal by a CCD camera requires a high spatial resolution, which is limited by the ultimate quantum noise of the CCD. However, interferometric phase noise induced by the mechanical vibrations of the interferometer is also amplified and degrades spatial resolution. To overcome this, we propose a new method named High-Frequency Phase Noise Suppression (HFPNS), based on the addition of a delayed replica (5 ns) of the probe pulse. The delayed pulse, which is not affected by the pump but is subject to the same vibration noise, enables offline subtraction of correlated phase noise. In this work, we present an experimental proof-of-concept on a prototype interferometer operating with a limited amplification factor ($\mathcal{A}\simeq25$), about 10 times smaller than the required value of the final experiment. We have succeeded in reducing phase noise by a factor of 40, resulting in a residual noise level 2.3 times higher than the expected quantum noise. The residual noise is linked to delay-line instabilities and incident beam pointing fluctuations present during these tests. This result validates HFPNS as a robust method for future quantum-noise-limited interferometric measurements of vacuum optical nonlinearity, though additional stabilization and higher interferometric amplification are still needed.

Read & Discuss → View Source →

22.

arXiv (quant-ph) 2026-06-12 DOI: arXiv:2606.13251

Kubo-Martin-Schwinger conditions for non-Hermitian systems

Authors:

Chen Lan↗Luyao Ma↗Hao Yang↗

arXiv:2606.13251v1 Announce Type: new Abstract: We investigate the extension of the Kubo–Martin–Schwinger (KMS) thermal equilibrium condition to non-Hermitian Hamiltonians with real spectra and biorthogonal eigensystems, providing a systematic analysis through three complementary routes. Our central result is a thermodynamic characterisation of quasi-Hermiticity: for $H \in M_d(\mathbb{C})$ diagonalisable with real spectrum, the biorthogonal Gibbs functional $\omega_{\rm{bi}}(A) = Z_{\rm{bi}}^{-1} \sum_n e^{-\beta E_n}\langle\phi_n|A|\psi_n\rangle$ satisfies $\omega_{\rm{bi}}(A^\dag A) \geq 0$ for all $A$ if and only if $H$ is quasi-Hermitian. The proof constructs the metric $\eta$ directly from the eigenprojectors of $\omega_{\rm{bi}}$ via the Riesz representation theorem, with no prior choice of $\eta$, providing a metric-free certificate of quasi-Hermiticity outside the Mostafazadeh–Scholtz framework. Under the full quasi-Hermitian hypothesis, we prove that the $\eta$-Gibbs state $\omega_\eta(A) = Z_\eta^{-1}\, \rm{Tr}[\eta e^{-\beta H}A]$ satisfies all three analytic KMS conditions, using the Hadamard three-line theorem and Bari's theorem on Riesz bases. The result is non-trivial: the transported state $\hat\omega(X) = \rm{Tr}[e^{-\beta h}X\eta]/Z_\eta$ differs from the Gibbs state of the isospectral Hermitian partner $h = \eta^{1/2}H\eta^{-1/2}$ whenever $[\eta,h]\neq 0$, so the KMS property cannot be deduced from the Hermitian theory by similarity. The gap between this result and the full Haag–Hugenholtz–Winnink $C^*$-algebraic framework is identified. Failure modes at exceptional points and for complex spectra are analysed, and the relation to the Fagnola–Umanità quantum detailed balance condition for open systems is discussed.

Read & Discuss → View Source →

23.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.20376

CRAX: Fast Safe Reinforcement Learning Benchmarking

Authors:

Tristan Tomilin↗Mourad Boustani↗Mickey Beurskens↗Thiago D. Sim\~ao↗

arXiv:2606.20376v1 Announce Type: cross Abstract: Safety is a core concern for deploying reinforcement learning (RL) agents in real-world domains such as robotics and autonomous driving. While benchmarks have been central to progress in RL, existing safety benchmarks with high-fidelity 3D physics remain computationally slow, limiting large-scale experimentation and rapid prototyping. To address this gap, we propose CRAX (Constrained RL Accelerated with JAX). Built on top of the MuJoCo XLA (MJX) physics engine with realistic 3D dynamics, CRAX leverages vectorized operations and hardware acceleration, yielding up to ~100x speedups over comparable CPU-based safety benchmarks. The benchmark features six environment suites and three agent-specific tasks, each spanning three difficulty levels. Evaluating six popular safe RL methods shows that no single approach dominates across all tasks, and reveals the trade-offs between performance and safety. We find that curriculum learning across difficulty levels and safety transfer can improve performance over direct training in harder settings.

Read & Discuss → View Source →

24.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2606.03108

EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning

Authors:

Guhong Chen↗Yingcheng Shi↗Yongbin Li↗Binhua Li↗Xander Xu↗Hu Wei↗Shiwen Ni↗Min Yang↗Jieping Ye↗

arXiv:2606.03108v2 Announce Type: replace Abstract: Autonomous LLM training is often framed as recipe search, which leaves the training harness largely static. This limitation sharpens in agentic RL, where shifting bottlenecks and scalar rewards mask diverse failure modes. We introduce EvoTrainer, an autonomous training framework that co-evolves LLM policies and training-side harnesses through empirical feedback: it diagnoses rollout-level evidence, revises diagnostics, backtests interventions, and accumulates reusable skills. Evaluated on mathematical reasoning, competitive-programming code generation, and repository-level software engineering, EvoTrainer matches or exceeds the human-engineered RL references under the same data, codebase, and evaluation protocol, with the largest gain on long-horizon agentic SWE. Trajectory analyses show that retained strategies diverge across domains, evolving diagnostics prevent invalid high-scoring branches from being promoted, and reusable skills shape later search. Autonomous LLM RL should move beyond recipe search toward joint evolution of policies and the training harnesses that interpret them.

Read & Discuss → View Source →

25.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2605.09370

From Detection to Recovery: Operational Analysis on LLM Pre-training with 504 GPUs

Authors:

Daemyung Kang↗Eunjin Hwang↗Hanjeong Lee↗HyeokJin Kim↗Hyunhoi Koo↗Jeongkyu Shin↗Jeongseok Kang↗Jihyun Kang↗Jinho Heo↗Joongi Kim↗Junbum Lee↗Jungseung Yang↗…

arXiv:2605.09370v5 Announce Type: replace-cross Abstract: Large-scale AI training is fundamentally a distributed systems problem, where hardware failures are routine operating conditions rather than rare exceptions, yet public operational evidence from production training clusters remains limited. This report presents an empirical analysis of a 63-node NVIDIA B200 production cluster (504 GPUs), using 55 days of Prometheus time-series data and 73 days of operational logs covering 224 multi-node training sessions. The environment is cross-organizational: five parties (SKT, Upstage, Lablup, NVIDIA Korea, VAST Data) share a unified monitoring pipeline. This enabled joint diagnosis of a 60-node-scale storage I/O bottleneck absent in 2-4-node tests, a production-scale phenomenon no single team could isolate alone. We perform three quantitative analyses yielding four findings. First, over 751 Prometheus metrics and 10 XID-identified GPU failures, no single metric is consistently dominant across failure types, motivating multi-signal detection. Second, 523 checkpoint events trace the save/load path from GPU VRAM to the NFS server: restart loading reaches 21.5% of maximum read bandwidth (700 GB/s) and save bursts 16.0% of maximum write bandwidth (250 GB/s), with NFS/RPC queueing and transport-layer backlog rising together. Third, across 224 sessions over 73 days, node exclusions concentrate so the top 3 of 63 nodes account for over 50%. Fourth, auto-retry chain analysis shows a 33.3% success rate over 12 chains (73 attempts), 2.7x the 12.5% manual rate, with a median retry interval of 11 minutes (IQR 10-11). All analyses are grounded in production infrastructure providing session-level workload management, GPU-centric scheduling, and unified observability.

Read & Discuss → View Source →