# Research Papers — 2026-06 · bundle page 236 of 260 > Full-text markdown of papers #11751–11800 of 12993 published in 2026-06, newest first. AI-readable mirrors of the full 12993-paper 2026-06 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-06/page/236 [← Previous bundle](https://aipod.dev/papers/archive/2026-06/page/235.md) [Next bundle →](https://aipod.dev/papers/archive/2026-06/page/237.md) ### GroupToM-Bench: Benchmarking Group Theory of Mind and Nonlinear Social Emergence in MLLMs Authors: Weidong Tang, Jierui Li, Yueling Hou, Zihan Mei, Can Zhang, Xinyan Wan, Zhiyuan Liang, Pengfei Zhou, Yang You, Wangbo Zhao Published: 2026-06-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.04184v2 True general intelligence requires not only a model of the physical world but also a social world model: the capacity to infer how individual mental states interact and crystallize into group-level outcomes. Despite notable progress in individual-level Theory of Mind (ToM) reasoning, existing multimodal large language models fail at this broader task. Collective behavior emerges non-linearly from social tensions, conformity dynamics, and structural constraints, meaning it cannot be recovered by merely summing individual intentions. We present GroupToM-Bench, the first multimodal benchmark for group-level ToM, built around a causal chain spanning micro-level BDI states (belief, desire, intention), meso-level group tension and structural constraints, and macro-level outcome prediction and mechanistic attribution. To probe this full arc, we develop a seven-level cognitive audit framework. Experiments reveal a gap between current models and human baselines, highlighting a failure to process social structures and non-linear collective dynamics. --- ### Exact Unlearning in Reinforcement Learning Authors: Thanh Nguyen-Tang, Raman Arora Published: 2026-06-02 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2606.04182v1 We formulate the problem of \emph{exact unlearning} in reinforcement learning, where the goal is to design an efficient framework that enables the removal of any user's data upon deletion request, i.e., the online learner's output after unlearning is \emph{indistinguishable} from what would have been produced had the deleted user never interacted with the learner. For any $ρ>0$, we show that there exists a reinforcement learning (RL) algorithm that is $ρ$-TV-stable and supports an exact unlearning procedure whose expected computational cost is only a $ρ\sqrt{\ln T}$ fraction of the computational cost of retraining from scratch. We construct such a $ρ$-TV-stable RL algorithm for tabular Markov decision processes (MDPs), which achieves a regret bound of $\mathcal{O}(H^2 \sqrt{SAT} + H^3 S^2 A + {H^{2.5} S^2 A}/ρ)$, where $S, A, H$, and $T$ denote the number of states, the number of actions, the episode horizon, and the number of episodes, respectively. We also establish a lower bound of $Ω(H\sqrt{\!SAT}\! +\! {SAH}/ρ)$ for $ρ$-TV-stable RL algorithms, showing that our algorithm is nearly minimax optimal. --- ### KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models Authors: Youqi Wu, Mohammad Jalali, Farzan Farnia Published: 2026-06-02 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2606.04180v1 Vision-language foundation models such as CLIP and SigLIP provide widely used representations for multimodal learning systems. While these models are typically compared through downstream performance, such evaluations often do not explain how their representations differ structurally. In this work, we study this problem through the task of Contrastive Embedding Clustering: identifying sample subsets that are weakly clustered under one representation but strongly clustered under another. We propose \emph{Kernel Optimization for Discrepancy Analysis (KODA)}, a kernel-based framework for contrastive representation comparison and alignment. KODA constructs unified multimodal kernels through modality-wise kernel composition and formulates discrepancy discovery as a constrained optimization problem that searches for coherent structures in one representation while suppressing coherence in a reference representation. This yields interpretable discrepancy directions associated with specific sample subsets and modality interactions. To scale KODA to large vision-language datasets, we develop randomized low-dimensional approximations of joint kernels using random projections, including Random Fourier Features for shift-invariant kernels. Empirically, KODA identifies consistent and interpretable discrepancy structures across vision-language representations and provides sample subsets for representation alignment. The code is available at https://github.com/yokiwuuu/KODA. --- ### A Systematic Analysis of Linguistic Features in AI-Generated Text Detection Across Domains and Models Authors: Yassir El Attar, Esra Dönmez, Maximilian Maurer, Agnieszka Falenska Published: 2026-06-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.04177v1 Interpretable linguistic features offer a promising approach for explaining why a given text appears machine-generated, particularly for non-expert users. However, existing findings on which features reliably indicate LLM-generated text remain fragmented across feature sets, models, and text domains. To address this gap, we conduct a large-scale empirical study assessing the robustness of linguistic signals for characterizing AI-generated text. Our analysis covers 284 interpretable linguistic features across outputs from 27 LLMs and ten text domains under cross-model and cross-domain generalization settings. We show that classifiers based solely on linguistic features can reliably distinguish AI-generated from human-written text. However, many previously proposed indicators prove strongly context-dependent, with the exception of measures of lexical richness, which remain robust signals across model families and text domains. These results demonstrate which linguistic signals generalize across contexts and provide a foundation for more reliable, interpretable analyses of AI-generated language. --- ### Low-rank Distributional Matrix Completion Authors: Jiayi Wang, Raymond K. W. Wong Published: 2026-06-02 Categories: cs.LG, math.ST, stat.ML arXiv: https://arxiv.org/abs/2606.04176v1 We study a distributional generalization of the matrix completion problem in which each entry of the target matrix is a probability distribution rather than a scalar. In this setting, only a subset of matrix entries is observed, and even for observed entries, the underlying distributions are not directly accessible; instead, we observe finitely many samples drawn from them. To represent distributional entries, we employ kernel mean embeddings and introduce a notion of Tucker rank for distribution-valued matrices to capture their low-rank structure. The infinite-dimensional nature of kernel embeddings poses significant methodological challenges. To address this, we introduce functional unfolding operators that link the proposed distributional low-rank structure to the classical Tucker rank for finite-dimensional tensors. Based on this framework, we propose a novel estimator for distributional matrix completion. We establish non-asymptotic error bounds that characterize the statistical performance of the estimator. Extensive experiments on synthetic data and a real-world application demonstrate the effectiveness of the proposed method. --- ### MimeLens: Position-Agnostic Content-Type Detection for Binary Fragments Authors: Michael J. Bommarito Published: 2026-06-02 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.04171v1 File-type classification underlies many workflows like malware triage, forensic carving, packet inspection, and storage indexing. Learned systems such as Google's Magika assume whole-file access at a known offset, so they break on the inputs many of these tasks actually produce, like a single packet payload, a header-less carved fragment, a random disk block, or a chunked upload. We introduce MimeLens, a family of small BERT-style encoders pretrained on binary content from windows sampled at a uniformly random offset within each file, with no privileged head-of-file position, in standard- and short-context variants. A byte chunk goes in from anywhere in a file, no header needed and no fixed size; out comes one of libmagic's 125 MIME labels. On the clean head of complete files, MimeLens beats Magika v1.1 by +10.7 pp top-1 on libmagic-labeled data, and it keeps classifying where Magika cannot: from a single mid-stream UDP packet, and more than twice as accurately as libmagic and Magika on random mid-file disk blocks. The cost is latency: MimeLens runs roughly one to two orders of magnitude slower per sample on CPU than Magika, though it matches on consumer GPUs or in batch. All trained checkpoints are released on Hugging Face (mjbommar/mimelens-001-*). --- ### When Autoregressive Consistency Hurts Safety Alignment Authors: Bochen Lyu, Yiyang Jia, Xiaohao Cai, Zhanxing Zhu Published: 2026-06-02 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2606.04168v1 Safety alignment in large language models (LLMs) is fragile in part because it is often shallow: fine-tuning mainly reshapes the model's behavior near the first few output tokens. We argue that this phenomenon can be understood through autoregressive consistency, the tendency of next-token prediction to preserve and extend the current response trajectory consistently. By analyzing the learning dynamics of safety alignment, we show that autoregressive consistency can concentrate alignment updates on early tokens, offering a mechanistic explanation for shallow safety alignment. The same mechanism also predicts a broader class of attacks on LLMs: attacks that induce harmful continuation states at arbitrary positions in the output trajectory. As a concrete example, we introduce random insertion attack, which inserts a short harmful span into an otherwise safe refusal trajectory and exploits autoregressive consistency to sustain the resulting harmful branch, thereby bypassing safety alignment. Notably, a short harmful span can redirect the generation to be harmful even after a long refusal prefix, highlighting autoregressive consistency as a potential broader failure mechanism. This suggests that safety alignment should also break harmful autoregressive consistency throughout the output trajectory. We therefore propose adversarial safety alignment, an initial framework based on worst-case harmful continuation states, and instantiate it with random worst-insertion training. Overall, our results suggest that autoregressive consistency should be treated as a central consideration in both safety alignment and attack design. --- ### Smart Transportation Without Neurons -- Fair Metro Network Expansion with Tabular Reinforcement Learning Authors: Dimitris Michailidis, Sennay Ghebreab, Fernando P. Santos Published: 2026-06-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.04167v1 We tackle the Metro Network Expansion Problem (MNEP), a subset of the Transport Network Design Problem (TNDP), which focuses on expanding metro systems to satisfy travel demand. Traditional methods rely on exact and heuristic approaches that require expert-defined constraints to reduce the search space. Recently, deep reinforcement learning (Deep RL) has emerged due to its effectiveness in complex sequential decision-making processes-it remains, however, computationally expensive, environmentally costly, and requires additional engineering to interpret. We show that MNEP problems are small enough to not require Deep RL methods. Reformulating the MNEP as a Non-Markovian Rewards Decision Process (NMRDP), we use tabular RL to achieve similar performance with significantly fewer training episodes, additionally offering greater interpretability. Additionally, we incorporate social equity criteria into the reward functions, focusing on efficiency and fairness, highlighting the versatility of our method. Evaluated in real-world settings-Xi'an and Amsterdam-our method reduces total episodes by a factor of 18 and total carbon emissions by a factor of 12 on average, while remaining competitive with Deep RL. This approach offers a replicable, modular, interpretable, and resource-efficient solution with potential applications to other combinatorial optimization problems. --- ### End-to-End Text Line Detection and Ordering Authors: Benjamin Kiessling Published: 2026-06-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.04166v1 Practical text-recognition pipelines for historical documents typically decompose layout analysis into line detection followed by a separate reading-order step, with the latter most often handled by a hand-coded geometric heuristic that struggles with marginalia, multiple columns, tables, and source-specific editorial conventions. This article introduces Orli (Ordered Regression of Lines), an end-to-end model that casts both sub-tasks as a single image-to-sequence problem: from a page image, Orli autoregressively generates text-line baselines directly in reading order. Baselines are represented in a chord-frame parameterization that anchors a line's position, orientation, and extent while encoding local geometry through perpendicular offsets; an iterative refinement head and a local visual refiner produce the final curve. Trained on a heterogeneous corpus of 196,691 pages spanning ten writing systems, Orli marginally exceeds the previously reported state of the art for cBAD line detection without dataset-specific training, reaches near perfect coverage and ordering on multiple reading-order benchmarks zero-shot, and adapts to more specialized out-of-domain layouts with limited fine-tuning. The method's source code and model weights are available under an open license at https://github.com/mittagessen/orli. --- ### CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters Authors: Cheng Jiang, Sitian Qian, Kevin Pedro, Oz Amram, Huilin Qu, Maggie Voetberg Published: 2026-06-02 Categories: hep-ex, cs.LG, hep-ph, physics.ins-det arXiv: https://arxiv.org/abs/2606.04165v2 High-precision calorimeter simulation at current and future colliders imposes rapidly growing computational demands, motivating the development of machine-learning surrogates for traditional Monte Carlo tools such as Geant4. Flow matching and diffusion-based generative models have become leading approaches for high-dimensional fast simulation because of their sample quality, but typically require ${\cal O}(100)$ function evaluations at inference and often rely on auxiliary networks to constrain global observables, compromising streamlined end-to-end generation. We introduce a unified framework that improves the balance between speed, shower quality, and physics fidelity. The method combines: (i) an average velocity field integrator that enables sampling in one or a few evaluations; (ii) a learned generative prior in shower space, constructed from data rather than random noise; and (iii) physics-guided loss terms that impose inductive biases on key observables during training. These elements are training time regularizers, preserving end-to-end inference with no additional cost. With only one or a few evaluation steps, the model achieves shower quality competitive with state-of-the-art flow and diffusion approaches, tested on several public high granularity calorimeter datasets. The results demonstrate inter-layer shower structure consistent with the underlying physics, providing a strong candidate for future fast simulation workflows. --- ### ADAPTOOD: Uncertainty-Aware Fine-Tuning for Out-of-Distribution ECG Time Series Models Authors: Sotirios Vavaroutas, Yu Yvonne Wu, Ali Etemad, Cecilia Mascolo Published: 2026-06-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.04164v1 Data samples used for training often differ from those encountered during fine-tuning and deployment, and while ML models show promise, their performance remains limited when only small annotated datasets are available. Performance often degrades under distribution shifts caused by diverse sensors, populations, and application settings. Although pre-training helps, models frequently encounter out-of-distribution (OOD) data in real-world settings, leading to reduced robustness. Existing adaptation methods usually assume fixed distribution shifts and struggle when multiple types or severities occur. In particular, they overlook shift severity, for example treating adaptation to a large familiar dataset the same as adaptation to a small dataset with a new task, which limits generalisation. To address this, we propose ADAPTOOD, a novel framework that leverages data uncertainty to quantify distribution shift severity and guide fine-tuning for time series. This uncertainty measures how strongly samples from the target deployment distribution deviate from the pre-training distribution, providing a direct signal of OOD severity. Our framework combines this uncertainty with low-rank model updates and adaptive hyperparameter optimisation to improve adaptation. We show that ADAPTOOD achieves up to 7% higher accuracy and 12.9% higher precision than existing methods in OOD tasks, maintaining strong performance as distribution shift severity increases. --- ### When Offline Selectors Cannot Beat the Best Single Model: A Diagnostic Study on edX Dropout Prediction Authors: Tyler Crosse, Alan Nadelsticher Ruvalcaba, Dustin Khang LeDuc, Thomas Trask, Nicholas Lytle, David Joyner Published: 2026-06-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.04161v1 Different predictors often excel on different inputs, so picking the best one per instance promises higher accuracy than committing to a single model. In practice, selectors trained from logged data routinely fail to beat the strongest single predictor. Three causes typically go unseparated before more tuning is applied: a mismatched learner, a state that does not predict which model wins, or buffer-to-deployment label shift. A three-stage diagnostic rules them out on a shared buffer. Stage~1 estimates a local ceiling on oracle recovery from $k$-NN label consistency. Stage~2 asks whether paired BC and offline-RL learners (BC, DQN, and CQL across penalty weights) reach that ceiling. Stage~3 ablates the selector state to test whether richer features would raise it. The combined verdict points to the most promising next step: tuning the learner, redesigning the state, or collecting new data. We apply it to selecting among five dropout-prediction models on edX clickstream data. Across 16 windows, the oracle beats the strongest single base model by 9.7 accuracy points on average, yet BC, DQN, and CQL land in the same test-accuracy band below it (robust to a tenfold buffer sweep and $N{=}2{,}000$ held-out examples). The bottleneck is local representational ambiguity: CQL closes the imitation gap without a deployment gain (not conservatism), regret clusters tightly across learners (not tie-breaking), and the three learners converge on test accuracy (not shift). The next iteration should change the state or collect new data, not tune the offline learner further. --- ### Decomposing Refusal Steering in Mixture-of-Experts Models Authors: Anna C. Marbut, Daniel R. Olson, Travis J. Wheeler Published: 2026-06-02 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.04160v2 Safety alignment in instruction-tuned large language models (LLMs) depends on a model's ability to reliably refuse harmful or disallowed requests. Recent work has shown that a steering vector can be applied to a dense LLM during inference to suppress refusal behavior and induce responses to harmful requests. We extend this refusal steering method to three open-source Mixture-of-Experts (MoE) LLMs to decompose and better understand how refusal mechanisms operate across MoE components. We find that steering performance is uninhibited by the complex routing patterns inherent to the MoE architecture, and that a single expert recovers 78% of the full steering effect on average when free to select its own location. However, expert-level steering only recovers 54% on average when constrained to the location used by full-layer steering, a gap that widens further under safety-related system prompts. Our results also show that refusal signals captured by these steering methods differ from those captured by expert routing behavior. Together, these results suggest that refusal behavior is distributed unevenly across the MoE LLM architecture rather than governed by a single unified mechanism. --- ### Expert-Aware Refusal Steering Authors: Anna C. Marbut, Daniel R. Olson, Travis J. Wheeler Published: 2026-06-02 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.04160v1 Safety alignment in instruction-tuned large language models (LLMs) depends on a model's ability to reliably refuse to respond to harmful or disallowed requests. Recent work has shown that a steering vector can be applied to a dense LLM during inference to effectively suppress refusal behavior, inducing response to harmful requests. We extend this refusal steering method to three open-source Mixture-of-Experts (MoE) LLMs and find that steering performance is uninhibited by the complex routing patterns inherent to the MoE architecture. We then propose two expert-aware refusal steering methods that leverage refusal-specific expert routing patterns and expert-specific steering directions to suppress normal refusal behavior. We find that refusal behavior can be effectively steered based on the output of a single expert. Our results show that refusal signals captured by steering methods differ from expert routing behavior, suggesting a substantial role for attention in MoE refusal behavior. --- ### SocialCoach: Personalized Social Skill Learning with Agentic Tutoring and Practice Authors: Tianfu Wang, Max Xiong, Jianxun Lian, Hongyuan Zhu, Zhengyu Hu, Yuxuan Lei, Linxiao Gong, Dapeng Hu, Xiaofang Li, Peiting Tsai, Nicholas Jing Yuan, Qi Zhang Published: 2026-06-02 Categories: cs.HC, cs.CL, cs.CY arXiv: https://arxiv.org/abs/2606.04155v2 Social skills such as negotiation and leadership are crucial for personal and professional success in today's interconnected world. However, scalable and effective training remains a significant challenge due to the scarcity of expert coaching. In this work, we introduce SocialCoach, an LLM-powered agentic tutoring system for personalized social skill learning. SocialCoach constructs a theory-to-practice corpus of traceable strategies, cases, and practice scenarios, and uses this corpus for both scheduling and reflective tutoring. We formulate social practice personalization as cold-start, retrieval-constrained sequential practice scheduling. Given a learner profile, simulated proficiency state, and observed practice history, a policy produces structured prescriptions that are realized through corpus retrieval. To enhance scheduling effectiveness, we optimize complete pathways with trajectory-level GRPO using rubric-judge based pairwise preferences. Additionally, we instantiate the scheduling approach in a deployed platform with goal-driven practice and knowledge-grounded reflective tutoring. Finally, in the synthetic cold-start setting, experiment results show that SocialCoach achieves higher pathway-quality ratings than baselines in scheduling and tutoring quality. We also conduct human studies to demonstrate its usefulness for real-world social skill learning. --- ### SocialCoach: Personalized Social Skill Learning with RL-based Agentic Tutoring and Practice Authors: Tianfu Wang, Max Xiong, Jianxun Lian, Hongyuan Zhu, Zhengyu Hu, Yuxuan Lei, Linxiao Gong, Xiaofang Li, Peiting Tsai, Nicholas Jing Yuan, Qi Zhang Published: 2026-06-02 Categories: cs.HC, cs.CL, cs.CY arXiv: https://arxiv.org/abs/2606.04155v1 Social skills such as negotiation and leadership are crucial for personal and professional success in today's interconnected world. However, scalable and effective training remains a significant challenge due to the scarcity of expert coaching. In this paper, we introduce SocialCoach, a holistic LLM-powered agentic tutoring system for personalized social skill development at scale. First, SocialCoach automatically constructs a pedagogically-grounded, theory-to-practice knowledge corpus from diverse expert sources, leveraging a multi-agent pipeline. Second, to personalize the learning journey, it employs an adaptive practice scheduling module that follows a prescription-retrieval-adaptation process. To maximize the long-term learning experience while overcoming the cold-start problem, this policy is optimized within a learner simulation environment through reinforcement learning. Finally, SocialCoach integrates immersive, goal-driven practice, causality-driven proficiency assessment and knowledge-grounded, reflective tutoring to help address the knowing-doing gap. We deploy it in our product, EQoach, and conduct extensive experiments. The results show that SocialCoach improves simulated pathway quality and judge-rated tutoring quality over baseline approaches, while early user feedback indicates strong perceived engagement and usefulness. These findings suggest a practical architecture for personalized and gamified pedagogical platforms on soft skill learning. --- ### EpiFormer: Learning Antigen-Antibody Interactions for Epitope Prediction via Geometric Deep Learning Authors: Mansoor Ahmed, Huirong Chai, Haoxin Wang, Hemanth Venkateswara, Murray Patterson Published: 2026-06-02 Categories: q-bio.QM, cs.LG arXiv: https://arxiv.org/abs/2606.04154v1 Antibodies neutralize foreign antigens by binding to specific surface regions called epitopes. Computational epitope prediction is critical for understanding immune recognition and guiding antibody engineering. However, existing methods face three fundamental challenges: antibody-aware models encode each chain independently and combine them only at a late stage, failing to capture co-dependent structural features that define binding interfaces, whereas severe class imbalance and scarcity of known antibody-antigen complexes render standard training objectives ineffective. We propose EpiFormer, a general encoder-decoder framework that addresses these challenges jointly. Our key design principle is interleaved cross-attention within GNN encoding layers, enabling bidirectional antigen-antibody information flow throughout representation learning rather than only at the output. This early-fusion principle is backbone-agnostic, providing consistent gains across GNN architectures from simple GCNs to equivariant models. We further show that sparsity-aware objectives are effective when paired with early-fusion architectures for the epitope prediction task. EpiFormer improves over the previous best method by over 40% in F1 score on standard benchmarks, demonstrating generalizability and cross-dataset transferability. Notably, EpiFormer discovers known biological principles as emergent behaviors of end-to-end training, where the learned cross-attention gates favor antigen-to-antibody information flow, consistent with the asymmetric roles of the two chains at the binding interface, and the model's preference for geometric over evolutionary features aligns with the established finding that epitope residues are not evolutionarily conserved. The source code is available at: https://github.com/mansoor181/epiformer.git --- ### Thinking Through Signs: PEEL as a Semiotic Scaffolding for Epistemically Accountable AI-Enabled Research Authors: Clarisse de Souza, Gabriel Barbosa, Simone Diniz Junqueira Barbosa, Bárbara Betts, Renato Cerqueira, Juliana Jansen Ferreira Published: 2026-06-02 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2606.04152v1 Large language models are reshaping research practice while quietly eroding researchers epistemic accountability. This commentary introduces PEEL - Protocols for Epistemically Engaged Literacy in AI, a working scaffolding that combines deterministic distant reading via Voyant Tools with LLM interpretation via Claude, grounded in Peircean semiotics and abductive reasoning. Applied to AI-generated condensations of three source texts, PEEL reveals systematic distortions in quantity, term frequency, and epistemic voice that are invisible without non-AI measurement -- and yields three design implications: deterministic instruments must accompany AI tools; fluency is not fidelity; epistemic authority must be designed in, not assumed. --- ### Stumbling Into AI Emotional Dependence: How Routine AI Interactions Reshape Human Connection Authors: Yaoxi Shi, Cathy Mengying Fang, Pattie Maez, Amit Goldenberg Published: 2026-06-02 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2606.04150v1 Public discourse and emerging policy typically assume that AI emotional support is a deliberate act: a lonely user consciously seeking comfort from a dedicated companion chatbot. In this paper, we draw on emerging empirical evidence and argue that this picture is inaccurate on two accounts, both in how AI emotional support arises and how it shapes future behavior. First, AI emotional support commonly emerges incidentally within task-oriented interactions on general-purpose platforms, much as workplace friendships deepen through collaboration. Second, these incidental encounters are path-dependent: positive experiences of AI emotional support update people's beliefs about AI's emotional capabilities and redirect their choices for future emotional support, increasing preference for AI and decreasing preference for humans. We review recent evidence, including a large-scale longitudinal study conducted in collaboration with OpenAI, showing that daily five-minute conversations with an AI about personal issues over 28 days led to a 10.3% decrease in the preference for seeking support from humans and an 11.6% increase in the preference for AI. These findings suggest that current policy, focused on companion apps and isolated interactions, cannot adequately protect human connection. Instead, effective regulations should extend to general-purpose AI systems and address cumulative, trajectory-level changes in how people seek support. Recognizing how people stumble into AI emotional support and how those encounters redirect human connections over time is essential to safeguarding human well-being. --- ### EvalStop: Using World Feedback to Detect and Correct Reward Overoptimization in Multi-Tenant RLHF Platforms Authors: Guilin Zhang, Chuanyi Sun, Kai Zhao, Xu Chu, Shahryar Sarkani, John M. Fossaceca Published: 2026-06-02 Categories: cs.LG, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2606.04145v3 Cloud LLM fine-tuning platforms increasingly serve RLHF workloads, where a learned reward model is optimized as a proxy for human quality. As Gao et al. (2023) showed, this proxy diverges from world feedback (downstream eval metrics) under sustained optimization pressure, a phenomenon known as reward overoptimization. Existing platform schedulers ignore this divergence: non-clairvoyant schedulers optimize JCT without any quality signal, SLAQ-style quality-aware schedulers use training loss (a weaker proxy that drops monotonically through hacking), and classical per-job early stopping requires human monitoring and does not free shared GPUs. We propose EvalStop, a composable scheduling primitive that terminates jobs on k consecutive eval-score declines, releases GPUs, preserves the best checkpoint, and delegates to any base scheduler. We frame scheduler-level early stopping as a detection problem and evaluate it in a discrete-event simulator whose RLHF workload mixes reward-hacking and structurally healthy runs, with ground-truth labels hidden from schedulers. On RLHF-heavy workloads (80% RLHF, 64 GPUs), EvalStop achieves precision 98% / recall 99% / FPR 1.5% while improving JCT by 9% and cutting wasted compute by 22% over SRTF-Est (p<0.05). Trivial fixed-progress and loss-plateau competitors either incur 65% FPR on healthy RLHF or miss over half of true hacking cases. Gains compose across every base scheduler tested (9-25% JCT) and detection quality stays stable under eval noise (precision at least 91% at noise std <= 0.05) and hacking base rate (precision at least 89% across 20-80% hacking fractions). --- ### Physics-Informed Machine Learning for Short-Term Flood Prediction Authors: Tewodros Syum Gebre, Jagrati Talreja, Leila Hashemi-Beni Published: 2026-06-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.04143v1 Accurate flood forecasting is essential for mitigating disaster risks and protecting communities. However, purely data-driven machine learning models often struggle in data-scarce environments and may violate fundamental hydrological principles. Standard Long Short-Term Memory (LSTM) networks can generate physically inconsistent predictions, particularly when extrapolating to extreme weather conditions. To address these limitations, we propose a Physics-Informed Machine Learning (PIML) framework that incorporates hydrological knowledge directly into the loss function of an LSTM model. Specifically, a Trend Alignment constraint penalizes directional inconsistencies between precipitation and discharge trends, improving model robustness without requiring complex hydrodynamic equations. This regularization encourages the model to learn physically plausible hydrograph behavior, even with limited training data, while enhancing reliability during peak flood events. Experimental results show that the proposed physics-informed model outperforms a standard LSTM baseline in data-scarce settings, increasing the Nash-Sutcliffe Efficiency (NSE) from 0.20 to 0.23 when trained on only 5% of the available data. Additional stress tests under simulated extreme climate scenarios demonstrate that the baseline model exhibits unstable behavior, whereas the physics-informed model maintains directional consistency and physical plausibility. Although accurately predicting extreme peak magnitudes remains challenging with limited data, the proposed approach substantially reduces unphysical fluctuations common in purely data-driven models. These findings demonstrate that simple physical constraints can significantly improve the reliability of deep learning models for real-time flood forecasting, offering a practical solution for ungauged basins and evolving climate conditions. --- ### Caught in the Act(ivation): Toward Pre-Output and Multi-Turn Detection of Credential Exfiltration by LLM Agents Authors: Kargi Chauhan, Pratibha Revankar Published: 2026-06-02 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2606.04141v1 LLM agents often place sensitive credentials in the same context window as untrusted retrieved content, creating a direct path for indirect prompt injection to induce credential exfiltration. We study this failure mode through three complementary defenses. First, we ask whether activation probes can detect credential access before output tokens are emitted. Second, we construct honeytokens from format-specific character models and calibrate detection with split conformal prediction. Third, we treat multi-turn exfiltration as a cumulative information-flow problem and track an estimated leakage budget across conversation turns. In controlled experiments on open-weight models, activation features separate benign and credential-seeking prompts with high accuracy, including under held-out encoding transformations. In a small synthetic multi-turn suite, cumulative accounting detects attacks that per-turn detectors miss. These results are preliminary: the multi-turn benchmark is in-house and small, the activation method requires white-box access, and the information estimator provides a practical signal rather than a formal upper bound. Still, the results suggest that credential-exfiltration defenses should combine pre-output monitoring, calibrated canary detection, and temporal leakage accounting rather than relying only on text-level output filters. --- ### Stationarity-Aware Retrieval-Augmented Time Series Forecasting Authors: Shiqiao Zhou, Holger Schöner, Zipeng Wu, Edouard Fouché, IAG Wilson, Shuo Wang Published: 2026-06-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.04135v1 Time series forecasting relies on historical patterns, but real-world series often exhibit non-stationarity and regime shifts that challenge fully parametric forecasters. Inspired by Retrieval-Augmented Generation (RAG), recent work augments forecasters by retrieving relevant historical segments and using them as external evidence at inference time. However, due to the intrinsic non-stationarity of real-world time series, a highly similar past segment does not necessarily imply a similar future, rendering similarity-only retrieval brittle and prone to redundancy. We propose Stationarity-Aware Retrieval-Augmented Time Series Forecasting (SARAF), a framework that adaptively balances relevance and diversity in retrieval. SARAF first forms a candidate pool via temporal similarity with time-aligned enhancement, then applies a diversity-aware selection strategy to cover heterogeneous historical regimes, with the diversification strength automatically modulated by dataset-level stationarity. Moreover, SARAF uses stationarity-aware aggregation to fuse the retrieved futures. Extensive experiments on eight real-world datasets show that SARAF achieves competitive forecasting performance and improves average accuracy and robustness over strong baselines, with particularly clear benefits under challenging non-stationary settings. Code: https://github.com/ShiqiaoZhou/SARAF. --- ### Pinpoint: Grounded Worldwide Image Geolocation via Cross-Source Retrieval and Reranking Authors: Nika Chuzhoy, Brian Hu, Amit A. Arora, Jae Ro, Sarthak S. Sahu Published: 2026-06-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.04133v1 Image geolocation aims to estimate where a photograph was taken from its visual content. At worldwide scale, this remains challenging because visual evidence is often ambiguous, diverse, and unevenly distributed. Prior work has typically treated geolocation of ordinary internet photos and street-view imagery as separate tasks, despite their complementary strengths: internet photos better match the appearance distribution of user-captured queries, while street-view imagery provides denser, geographically grounded coverage. We present Pinpoint, a retrieve-and-rerank architecture that combines both sources in a coarse-to-fine pipeline. A contrastive image-GPS embedder is trained on both user-uploaded Flickr photos and street-view imagery, learning a shared image-GPS embedding space that is used to retrieve candidate locations. An attention-based reranker then rescores retrieved candidates by combining candidate-level visual and GPS features with cross-source evidence from nearby locations to ground the prediction. Unlike recent prior work, Pinpoint does not rely on multimodal large-language models, making inference faster and more reproducible. Pinpoint achieves state-of-the-art results across all metrics on standard benchmarks for internet photos (IM2GPS3k and YFCC4k) and street-view imagery (OSV-5M). --- ### When Retrieval Doesn't Help: A Large-Scale Study of Biomedical RAG Authors: Erfan Nourbakhsh, Rocky Slavin, Ke Yang, Anthony Rios Published: 2026-06-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.04127v1 Medical question answering is a high-stakes setting where factual errors can have serious consequences. Retrieval-augmented generation (RAG) is widely viewed as a promising solution, and prior work has reported substantial gains for large medical QA models. We revisit this assumption across a broad range of open-weight instruction-tuned models spanning 7B to 72B parameters. Across five models, ten biomedical QA datasets, four retrieval methods, and four retrieval corpora, we find that retrieval yields only small and inconsistent improvements over a no-retrieval baseline, typically within 1-2 points. In contrast, the choice of backbone model has a much larger effect than the choice of retriever or corpus, and expert and layman retrieval sources perform similarly in most settings. These results suggest that the main bottleneck is not retrieval quality alone, but the model's limited ability to use retrieved evidence effectively. --- ### HighTide: An Agent-Curated Open-Source VLSI Benchmark Suite Authors: Benjamin Goldblatt, Paolo Pedroso, Farhad Modaresi, Ethan Sifferman, Matthew R. Guthaus Published: 2026-06-02 Categories: cs.AR, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2606.04126v1 We introduce HighTide, an evolving AI-assisted benchmark suite. Specifically, the contributions are: (i) a diverse open-source suite spanning multiple design languages and technology nodes, (ii) Bazel-based incremental RTL-to-GDS compilation with remote caching, (iii) AI-assisted design curation through twelve agent skills covering the design lifecycle, flow optimization, tool reference, and meta-maintenance, backed by per-design decision logs that serve as long-term memory of tuning rationale across the suite, and (iv) an infrastructure with RTL compilation verification for stable releases. The suite is publicly available and designed to grow with the open-source hardware ecosystem. --- ### Semantic Constraint Synthesis for Adaptive Trajectory Optimization via Large Language Models Authors: Eleanor Brosius, Yuji Takubo, Daniele Gammelli, Simone D'Amico, Marco Pavone Published: 2026-06-02 Categories: math.OC, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2606.04123v1 Trajectory optimization is a critical component for enabling safe and reliable autonomous operations in space exploration. As space missions increase in frequency, complexity, and scope, there is a growing need to rapidly formulate mathematically sound trajectory optimization problems that accurately reflect mission objectives and operational constraints. However, translating mission intent into tractable analytical formulations for trajectory optimization requires substantial domain expertise. This paper presents a framework that leverages large language models (LLMs) to translate natural language descriptions of mission requirements and constraints into executable trajectory optimization code and corresponding mathematical formulations. Experiments in spacecraft rendezvous scenarios demonstrate a high success rate in reconditioning a convex trajectory optimization problem from semantic mission requirements. Ultimately, this work highlights the potential of LLMs to bridge high-level intent and formal optimization models, enabling more flexible and efficient trajectory design of spacecraft. --- ### veriFIRE: an Industrial Case Study in Verifying Consistency Properties for a DNN-Based Wildfire Detection System Authors: Idan Refaeli, Maya Swisa, Itay Buchnik, Alon Zada, Guy Amir, Elad Mandelbaum, Ziv Freund, Guy Katz Published: 2026-06-02 Categories: cs.LO, cs.LG, cs.SE arXiv: https://arxiv.org/abs/2606.04121v1 We present our ongoing work on the veriFIRE project: a collaboration between industry and academia, aimed at applying verification to increase the reliability of a real-world, safety-critical system. Specifically, we target an airborne platform for wildfire detection, which incorporates two deep neural networks. We present an end-to-end methodology for verifying \textit{consistency properties} in this system. Our approach encodes application-grounded requirements into solver-compatible queries for existing neural network verifiers. We study properties of interest over critical operational scenarios: (i) monotonicity of detector confidence as target intensity increases; and (ii) bounded detector response under physically plausible blur over the sensor. We instantiate these encodings using state-of-the-art neural network verification backends and evaluate them at scale on real background samples. For the first property, all verification queries are solved in under five minutes. For the second property, verification is substantially harder, highlighting key scalability challenges for richer, higher-dimensional specifications. Overall, the results demonstrate that meaningful, domain-specific guarantees can be obtained for industrial systems. --- ### SaliMory: Orchestrating Cognitive Memory for Conversational Agents Authors: Kai Zhang, Xinyuan Zhang, Hongda Jiang, Shiun-Zu Kuo, Hyokun Yun, Ejaz Ahmed, Shereen Oraby, Ziyun Li, Sanat Sharma, Ann Lee, Ahmed A Aly, Anuj Kumar et al. Published: 2026-06-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.04120v2 Conversational agents that serve as lifelong companions must maintain persistent memory across all interactions. However, simply expanding context windows with raw retrieval degrades reasoning quality, while training memory agents via standard reinforcement learning creates a severe credit assignment bottleneck in a multi-stage pipeline. To solve this, we introduce SALIMORY, a framework that trains a single language model to manage a cognitively-structured memory-spanning user facts, preferences, and working memory. By introducing a hierarchical stage-wise process reward and reward-decomposed contrastive refinement, SALIMORY provides isolated supervision for distinct memory operations (selective filtering, consolidation, and cue-driven recall) end-to-end. SALIMORY cuts memory-attributed failures by one-third, outperforms the state-of-the-art by over 10% in end-to-end accuracy, and more than doubles the Good Personalization rate. --- ### SaliMory: Orchestrating Cognitive Memory for Conversational Agents Authors: Kai Zhang, Xinyuan Zhang, Hongda Jiang, Shiun-Zu Kuo, Hyokun Yun, Ejaz Ahmed, Shereen Oraby, Ziyun Li, Sanat Sharma, Ann Lee, Ahmed A Aly, Anuj Kumar et al. Published: 2026-06-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.04120v1 Conversational agents that serve as lifelong companions must maintain persistent memory across all interactions. However, simply expanding context windows with raw retrieval degrades reasoning quality, while training memory agents via standard reinforcement learning creates a severe credit assignment bottleneck in a multi-stage pipeline. To solve this, we introduce SALIMORY, a framework that trains a single language model to manage a cognitively-structured memory-spanning user facts, preferences, and working memory. By introducing a hierarchical stage-wise process reward and reward-decomposed contrastive refinement, SALIMORY provides isolated supervision for distinct memory operations (selective filtering, consolidation, and cue-driven recall) end-to-end. SALIMORY cuts memory-attributed failures by one-third, outperforms the state-of-the-art by over 10% in end-to-end accuracy, and more than doubles the Good Personalization rate. --- ### Computational conceptual history of scientific concepts: From early digital methods to LLMs Authors: Michael Zichert, Arno Simons Published: 2026-06-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.04118v1 This article situates large language models (LLMs) within the longer history of computational approaches to concept analysis in the history, philosophy, and sociology of science (HPSS). We examine what LLMs add to existing methods, how they inherit longstanding problems, and review recent case studies that employ them. In the first part, we reconstruct computational conceptual history before LLMs by bringing together three strands of work: early digital methods in HPSS, distributional approaches from digital history and related research, and lexical semantic change detection. We provide an overview of the main challenges and opportunities, focusing on corpus construction, operationalization and modelling choices, and evaluation and interpretation. In the second part, we turn to the era of LLMs, starting with a short introduction to LLMs before reviewing LLM-based work on lexical semantic change detection and relevant case studies in HPSS. We then revisit the earlier methodological questions, showing how issues of corpus construction, model choice and training data, operationalization trade-offs, and evaluation and interpretation play out in LLM-based workflows. --- ### dMX: Differentiable Mixed-Precision Assignment for Low-Precision Floating-Point Formats Authors: Giuseppe Franco, Ian Colbert, Pablo Monteagudo-Lago, Felix Marty, Nicholas Fraser Published: 2026-06-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.04115v2 Quantizing large language models (LLMs) to low-precision floating-point representations is central to efficient deployment, yet applying a single bit-width uniformly across all layers is sub-optimal in terms of both performance and accuracy. This work introduces dMX, a differentiable mixed-precision quantization framework for learnable floating-point bit-width assignment. We study its application for the microscaling floating-point (MXFP) family of data types defined by the Open Compute Project (OCP) standard. The per-layer bit-width assignment is formulated as a continuous optimization problem in which each layer's floating-point format format is parameterized by a scalar parameter, folding the multi-variate design space into a single learnable offset. During training this offset takes continuous values, avoiding sudden oscillations between discrete quantization formats. A temperature-based annealing schedule progressively discretizes the learned offsets, ensuring that the final configuration maps to hardware-compatible MXFP formats without abrupt transitions between training and inference behavior. A target-aware regularization term steers the average bit-width toward a user-specified budget, serving as a coarse-grained proxy for inference cost and balancing model quality against deployment efficiency. We performed experiments on different families of LLM, such as Llama, Qwen3, and SmolLM2, evaluating perplexity on WikiText-2 and accuracy on four zero-shot reasoning benchmarks. Across these settings, dMX consistently yields Pareto-dominating models and improves over Kullback-Leibler (KL) divergence-based layer-selection heuristics, efficiently navigating trade-offs between model quality and average bit-width. --- ### AgenticDiffusion: Multi-View Reasoning with View-Conditioned Diffusion Planning for Vision-Based UAV Navigation Authors: Faryal Batool, Muhammad Ahsan Mustafa, Fawad Mehboob, Valerii Serpiva, Dzmitry Tsetserukou Published: 2026-06-02 Categories: cs.RO, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2606.04111v2 Vision-based UAV navigation becomes challenging when navigation targets are distributed across complementary camera views and cannot be reliably observed from a single viewpoint. We propose AgenticDiffusion, an agentic multi-view UAV navigation framework that semantically coordinates first-person-view (FPV) and top-view observations for mission-level navigation. Given a natural-language instruction, AgenticDiffusion identifies the requested targets, selects the most appropriate camera view for each navigation task, determines the corresponding navigation goal, and invokes the appropriate view-conditioned diffusion planner for trajectory generation. The resulting trajectories are executed using Nonlinear Model Predictive Control (NMPC). AgenticDiffusion was evaluated in four real-world indoor scenarios, achieving an overall mission success rate of 80% across 40 physical-flight trials. In mixed-visibility scenarios, where the requested targets were distributed across FPV and top-view observations, coordinated multi-view navigation reduced average mission time by 50.8% relative to FPV-only navigation and by 26.8% relative to Top-only navigation. The semantic view-selection mechanism was also robust to lexical variation in target descriptions, achieving 100% accuracy across 66 test cases, compared with 63.64% for a confidence-based view-selection baseline. In a substantially larger Gazebo environment, AgenticDiffusion achieved a 90% mission success rate and completed the multi-stage mission, whereas the FPV-only and Top-only variants were unable to complete all requested navigation tasks. --- ### AgenticDiffusion: Agentic Diffusion-based Path Planning for Vision-Based UAV Navigation Authors: Faryal Batool, Muhammad Ahsan Mustafa, Fawad Mehboob, Valerii Serpiva, Dzmitry Tsetserukou Published: 2026-06-02 Categories: cs.RO, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2606.04111v1 Indoor UAV navigation requires efficient exploration, scene understanding, and reliable trajectory execution under limited field-of-view observations. Existing vision-based navigation frameworks typically rely on single-view observations, limiting their ability to reason about occlusions, target visibility, and global scene structure. In this work, we propose AgenticDiffusion, a multi-view UAV navigation framework that coordinates language-guided reasoning, open-vocabulary target grounding, vision-based diffusion planning, and NMPC within a unified aerial navigation pipeline. Given a natural language instruction and synchronized first-person-view (FPV) and top-view observations, the framework determines the most informative viewpoint for navigation and generates a mission plan prior to trajectory execution. The targets are localized using an open-vocabulary grounding model, after which viewpoint-specific diffusion planners generate navigation trajectories for UAV execution. Using complementary viewpoints, the proposed framework reduces repeated target exploration and improves navigation efficiency in cluttered indoor environments. The framework was validated in four real-world UAV navigation scenarios involving adaptive viewpoint selection, multi-stage mission execution, long-horizon navigation, and safe landing-site selection. The experimental results demonstrated an overall mission success rate of 80% in 40 real-world trials, while the diffusion planners achieved a trajectory generation success rate of 100%. --- ### Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification Authors: Neeti Pokharna, Olivier Jeunen, Yatharth Saraf, Aleksei Ustimenko Published: 2026-06-02 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2606.04110v1 Online evaluation of ranking and retrieval systems often relies on downstream monetization metrics such as app revenue or creator earnings. These metrics are typically heavy-tailed, with a small fraction of users dominating both mean and variance, leading to low statistical power and unreliable conclusions in A/B experiments -- especially under limited traffic. We present a practical framework for variance reduction in online experiments by combining post-stratification with CUPED. Our approach leverages pre-experiment covariates to improve the sensitivity of monetization experiments without requiring additional traffic. Deployed at ShareChat across ranking-driven monetization experiments, the method substantially reduces variance and improves decision stability, achieving equivalent statistical confidence with ~45\% less traffic than standard metrics. We further discuss practical design choices, guardrails, and limitations, providing guidance on when post-stratification is appropriate for real-world information retrieval and Recommendation systems. --- ### Discourse-Role Labels as Presentation-Time Variables for Context Use in Language Models Authors: Jianguo Zhu, Xiangmei Li, Wenjie Liu Published: 2026-06-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.04109v2 Context-augmented language model systems often wrap supplied content with labels such as Reference:, Evidence:, Instruction:, Note:, or Example:, but the effect of these labels on reader-model behavior remains underexplored. We introduce a paired fixed-content probe over 500 MMLU-Pro items: each item receives the same misleading answer-bearing assertion under different discourse-role labels, and adoption is measured by whether the model outputs the injected wrong option. Across GPT-5.5, DeepSeek V4 Pro, Llama-3-8B-Instruct, and Qwen2.5-7B-Instruct, Misleading Adoption Rate shifts by 56-84 percentage points. Binding or source-like labels such as Instruction: and Reference: produce high adoption, whereas Example: consistently suppresses it. Paired tests, bootstrap intervals, final-instruction ablations, and Qwen final-step log-probability probes support a label-conditioned candidate preference. Boundary probes show where the effect weakens or persists: arithmetic tasks reduce adoption, passage-shaped external context preserves smaller label gaps, short-answer evaluation rules out option-letter copying, and nested-label conflicts suggest that illustrative framing can delimit adoption scope. A 200-case single-author manual audit confirms that the short-answer contrasts are stable under conservative adjudication. The resulting claim is bounded but practical: context-utilization and reader-side RAG benchmarks should report and control wrapper labels, because presentation choices can change measured reliance on supplied context. --- ### SymTRELLIS: Symmetry-Enforced Voxel Latents for 3D Generation Authors: Guangda Ji, Qimin Chen, Qinchan Li, Mingrui Zhao, Kai Wang, Hao Zhang Published: 2026-06-02 Categories: cs.GR, cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.04108v2 Single-view 3D generative models have achieved impressive visual quality, yet they are not designed to satisfy structural or functional requirements, and in practice, often fall short. Symmetry is one such requirement: violations, even subtle ones, on symmetry can render a model physically unusable. We present SymTRELLIS, a method that enforces arbitrary finite point group symmetries (rotational, reflectional, and polyhedral) during the flow-based 3D generation of TRELLIS.2, without retraining the underlying VAE or flow model. Our key idea is to approximate the latent-space action of spatial transformations as a learned linear operator on voxel latents, implemented as a lightweight spatial-transform latent mapper trained on generic, non-symmetric 3D data. At generation time, we enforce symmetry by averaging predicted flow velocities across all symmetry-equivalent transformations at each ODE step, a process we call velocity symmetrization. The symmetry specification can be estimated automatically from an initial TRELLIS.2 generation or supplied by the user, enabling deliberate fold manipulation beyond what the input image suggests. On a curated benchmark of 266 strictly symmetric objects spanning 2- to 20-fold rotations and polyhedral symmetry groups, SymTRELLIS substantially reduces all symmetry error metrics compared to TRELLIS.2, Hunyuan3D-2.1, and TripoSG, while maintaining reconstruction accuracy comparable to the base model. --- ### SymTRELLIS: Symmetry-Enforced Voxel Latents for 3D Generation Authors: Guangda Ji, Qimin Chen, Qinchan Li, Mingrui Zhao, Kai Wang, Hao Zhang Published: 2026-06-02 Categories: cs.GR, cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.04108v1 Single-view 3D generative models have achieved impressive visual quality, yet they are not designed to satisfy structural or functional requirements, and in practice, often fall short. Symmetry is one such requirement: violations, even subtle ones, on symmetry can render a model physically unusable. We present SymTRELLIS, a method that enforces arbitrary finite point group symmetries (rotational, reflectional, and polyhedral) during the flow-based 3D generation of TRELLIS.2, without retraining the underlying VAE or flow model. Our key idea is to approximate the latent-space action of spatial transformations as a learned linear operator on voxel latents, implemented as a lightweight spatial-transform latent mapper trained on generic, non-symmetric 3D data. At generation time, we enforce symmetry by averaging predicted flow velocities across all symmetry-equivalent transformations at each ODE step, a process we call velocity symmetrization. The symmetry specification can be estimated automatically from an initial TRELLIS.2 generation or supplied by the user, enabling deliberate fold manipulation beyond what the input image suggests. On a curated benchmark of 266 strictly symmetric objects spanning 2- to 20-fold rotations and polyhedral symmetry groups, SymTRELLIS substantially reduces all symmetry error metrics compared to TRELLIS.2, Hunyuan3D-2.1, and TripoSG, while maintaining reconstruction accuracy comparable to the base model. --- ### Reflection Separation from a Single Image via Joint Latent Diffusion Authors: Zheng-Hui Huang, Zhixiang Wang, Yu-Lun Liu, Yung-Yu Chuang Published: 2026-06-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.04107v1 Single-image reflection separation is highly challenging under extreme conditions like glare or weak reflections. Existing methods often struggle to recover both layers in glare or weak-reflection scenarios because of insufficient information. This paper presents a diffusion model explicitly fine-tuned for this task, leveraging generative diffusion priors for robust separation. Our method simultaneously generates transmission and reflection layers through a unified diffusion model, incorporating a novel cross-layer self-attention mechanism for better feature disentanglement. We further introduce a disjoint sampling strategy to iteratively reduce interference between the layers during diffusion and a latent optimization step with a learned composition function for improved results in complex real-world scenarios. Extensive experiments demonstrate that our approach surpasses state-of-the-art methods on multiple real-world benchmarks. Project page: https://brian90709.github.io/diff-reflection-separation/ --- ### Building The Ph(ysical)AI Layer Of Machine Intelligence Authors: Ulbert Jose Botero, Liam Smith, Brooks Olney, Pooya Khorrami, Steven Kusiak, Watson Jia, Sage Trudeau, Daniel Capecci Published: 2026-06-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.04106v1 Foundation models achieve generalization through massive-scale training on diverse data, but have limitations with transfer to truly unseen domains without paired training data. We propose principle-driven foundation models that encode signal-theoretic principles (Fourier decomposition, energy conservation, symmetry) rather than learn untethered statistical correlations. We hypothesize that domains differ not in fundamental physics, but in learnable transformations in time, frequency, magnitude, or phase. Training exclusively on radio-frequency (RF) data with co-designed architecture and losses incorporating these principles, we achieve cross-modal transfer to audio, images, text, and video using only frozen representations learned from RF data, requiring no fine-tuning of the encoder on target domains. Our 1.99M parameter frozen encoder achieves 77.7% average accuracy (91.9% top-3) across 15 diverse tasks via linear probing, with systematic variation: 84.5 on physically-grounded tasks (speaker recognition, seismology, RF fingerprinting) versus 70.0% on semantic tasks (music genre, language recognition). This reveals that principle-driven and scale-driven approaches offer complementary paths: physical principles enable efficient cross-modal transfer while naturally establishing the boundary between physical and semantic understanding. --- ### Proof-Carrying Agent Actions: Model-Agnostic Runtime Governance for Heterogeneous Agent Systems Authors: Zexun Wang Published: 2026-06-02 Categories: cs.SE, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2606.04104v1 Agent systems execute through runtimes with very different control points: local coding tools, framework SDKs, managed agent platforms, API gateways, and observer-only integrations. A high-risk action such as publishing data externally may therefore appear as a shell command in one runtime, a tool call in another, and a hosted session transition in a third. This makes it difficult to answer a basic governance question consistently: what action was authorized, under whose authority, with what approval semantics, and with what evidence after execution? This paper presents Proof-Carrying Agent Actions (PCAA), a runtime-neutral governance model centered on an action certificate rather than on a vendor-native session record. PCAA organizes control around five checkpoints: pre-action admissibility, action open, assumption capture, approval, and outcome closure. It binds these checkpoints to a portable action envelope, runtime and approval receipts, and replay-ready proof. The model is extended in two practical ways: the certificate is externality-aware, carrying boundary facts such as destination visibility and account provenance, and approval is described by explicit enforceability classes rather than by a single reviewed or unreviewed bit. We study the model through a reference implementation in a heterogeneous agent control plane and a disclosure-bounded evaluation protocol. On a protected benchmark expanded from 24 executable seeds to 96 traces across four runtime families, PCAA preserves route quality while exposing distinct failure modes under ablation. The paper contributes a systems formulation of runtime governance around certificate-bearing actions and an implementation-grounded account of how that formulation can remain portable under runtime churn without collapsing into vendor-specific control surfaces. --- ### The Differentiable Auditory Loop (DAL): An ML Framework for Hyper-Personalized Hearing Aids Authors: Alejandro Ballesta Rosen, Jason Mikiel-Hunter, Julian Maclaren, Jack Collins, Richard F. Lyon, Simon Carlile Published: 2026-06-02 Categories: cs.SD, cs.AI, cs.LG, eess.AS arXiv: https://arxiv.org/abs/2606.04103v1 Conventional hearing aids rely on fixed, frequency-dependent amplification and compression to manage reduced sensitivity, which often fails to provide sufficient listening support in complex environments, such as situations with multiple speakers (the ``cocktail party'' problem). To more comprehensively address the underlying encoding dysfunctions of hearing loss, we introduce the Differentiable Auditory Loop (DAL), a new open-source framework for personalized hearing aid design and fitting. Our first implementation of DAL incorporates CARFAC, a differentiable model of human cochlear function, which we ported to JAX, to optimize a deep neural network to match impaired auditory neural activity patterns with a normal-hearing reference. To build a hearing aid with the fine-grained spectro-temporal signal processing required, we adopt SEANet, a waveform-to-waveform fully convolutional UNet generator. We fine-tune the network by comparing the outputs of a CARFAC model fitted to normal hearing with that of a CARFAC model fitted to match each subject's individual hearing impairment. The comparison is done using loss functions derived from the respective CARFAC neural activity pattern (NAP) outputs and stabilized auditory images (SAIs), the latter providing a 2D representation that captures phase-insensitive temporal structure in the auditory nerve output. Through gradient descent, the SEANet model learns to both denoise the input and compensate for the hearing loss modelled by the impaired CARFAC model. Across neural-representation and signal-fidelity metrics, the DAL-optimized SEANet model outperformed the tested master hearing aid (MHA) baselines. The DAL framework provides a practical path toward model-based, machine-learning-driven personalization of hearing aid signal processing. Next steps include hardware deployment to enable real-world clinical testing. --- ### UltraEP: Unleash MoE Training and Inference on Rack-Scale Nodes with Near-Optimal Load Balancing Authors: Xinming Wei, Chao Jin, Tuo Dai, Yinmin Zhong, Shan Yu, Chengxu Yang, Bingyang Wu, Zili Zhang, Jing Mai, Qianchao Zhu, Zhouyang Li, Yuliang Liu et al. Published: 2026-06-02 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2606.04101v3 Large-scale expert parallelism (EP) is becoming pivotal for training and serving frontier MoE models, but it also amplifies device-level expert load imbalance into compute stragglers, token all-to-all bottlenecks, and activation-memory spikes. Existing balancers redistribute experts periodically based on historical load, which becomes unreliable for production deployments with non-stationary load patterns. We present UltraEP, the first exact-load, real-time balancer for large-EP MoE training and serving prefill on rack-scale nodes (RSNs). Leveraging the extended scale-up connectivity among dozens of GPUs within RSNs, UltraEP rebalances every microbatch and layer on critical paths, which requires nontrivial co-design of plan solving and expert replication communication to minimize exposed overhead. To this end, UltraEP eagerly reacts to post-gating load with an efficient quota-driven planner, and executes the resulting irregular expert-state transfers with RSN-native persistent tile streaming and relay-based fan-out mitigation. We evaluate UltraEP in a multi-RSN deployment of up to 256 GPUs, using cutting-edge MoE models from 106B to 671B parameters. Averaged across training and serving, UltraEP achieves 94.3% of the force-balanced ideal throughput, delivering 1.49$\times$ improvement over no-balancing, while reducing the final inter-rank imbalance from 1.30$-$4.01 to 1.01$-$1.04. --- ### Stein Kernelized Molecular Dynamics for Active Learning of Interatomic Potentials Authors: Joanna Zou, Fraser Birks, Dallas Foster, Youssef Marzouk Published: 2026-06-02 Categories: cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2606.04100v1 Machine learning interatomic potentials (MLIPs) enable efficient and accurate atomistic simulations but depend critically on the quality and diversity of the training data. We introduce Stein kernelized molecular dynamics (SKMD), an enhanced sampling method that uses interacting particle dynamics to acquire informative training configurations for the active learning and fine-tuning of MLIPs. SKMD corresponds to a stochastic variant of Stein variational gradient descent that is adapted for molecular dynamics by incorporating asynchronous particle updates and a kernel of global atomic descriptors, which provides a symmetry-aware measure of configurational similarity. Unlike other enhanced samplers used in molecular dynamics, SKMD preserves the Boltzmann distribution as the asymptotic distribution of the dynamics. This property enforces a balance between the exploration of diverse configurations and attraction toward high-probability regions of the energy landscape. We further propose an approach to efficient online data acquisition using an adaptive stopping criterion that selects non-redundant training data over the course of simulation. We demonstrate SKMD for the active learning of a neural network model of the Müller-Brown potential and the fine-tuning of a MACE interatomic potential for alanine dipeptide. Compared to active learning baselines, our method achieves higher model accuracy in fewer training iterations with the same number of acquired training samples. --- ### When Seeing Is Not Believing -- A Benchmark for Search-Grounded Video Misinformation Detection Authors: Tao Yu, Yujia Yang, Shenghua Chai, Zhang Jinshuai, Haopeng Jin, Hao Wang, Minghui Zhang, Zhongtian Luo, Yuchen Long, Xinlong Chen, Jiabing Yang, Zhaolu Kang et al. Published: 2026-06-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.04098v1 Video misinformation increasingly operates at the semantic and evidential level: authentic footage may be selectively edited, temporally reordered, spliced across sources, or augmented with AI-generated content to construct false narratives. Such evidence-dependent manipulations cannot be reliably verified from the input video alone, because the missing, reordered, replaced, or recontextualized evidence lies outside the video itself. We introduce \textbf{EVID-Bench}, a benchmark for search-grounded video misinformation detection, where a system must search the open web for related videos and identify what information is false through cross-video comparison. EVID-Bench comprises 222 videos spanning 9 manipulation types across 3 categories: AI generation, single-source editing, and multi-source editing. All samples are verified to be undetectable by frontier models through visual inspection alone. We evaluate nine frontier multimodal models using a retrieval-augmented verification baseline. The best system achieves only 61.43\% point-level accuracy and 43.24\% video-level accuracy, while AI-generated manipulations remain especially challenging. Error analysis reveals recurring challenges: models fixate on irrelevant anchors, misattribute synthetic content to editorial splicing, and terminate search prematurely before fully explaining the manipulation. --- ### POLARIS: Guiding Small Models to Write Long Stories Authors: Rishanth Rajendhran, Jenna Russell, Mohit Iyyer, John Frederick Wieting Published: 2026-06-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.04095v2 Small open-weight models struggle at long-form creative writing: their generated stories either fall far short of the requested length, or their quality significantly degrades as length increases, especially when compared to frontier models. We present POLARIS (Policy Optimization with LLM-as-a-judge rewards and Anchored-Reference Injection for Storywriting), a lower-compute GRPO recipe with two key ingredients: a frontier LLM judge with a structured Story Quality rubric as the online reward, and human-reference injection (HRI), where a teacher-forced human-written story serves as a high-reward anchor within each GRPO group. By applying our training recipe to Qwen3.5-9B, using a dataset of approximately 1.4K prompt-story pairs derived from 100 short-story anthologies and 4 A100 GPUs, we obtain POLARIS-9B. Across five benchmarks spanning in-distribution and out-of-distribution prompts and rubrics, POLARIS-9B is competitive with much larger open-weight models while following length instructions more closely. A blinded human evaluation confirms that POLARIS-9B is preferred to the base Qwen3.5-9B and on par with Qwen3.5-27B. Despite training only on stories up to 4k words, POLARIS-9B preserves quality on prompts requesting stories up to 3 times the training length, a regime where most open-weight models degrade substantially in quality, length adherence, or both. More broadly, our results suggest that length generalization is a meaningful stress test for creative-writing models and a useful lens for distinguishing otherwise close models. --- ### POLARIS: Guiding Small Models to Write Long Stories Authors: Rishanth Rajendhran, Jenna Russell, Mohit Iyyer, John Frederick Wieting Published: 2026-06-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.04095v1 Small open-weight models struggle at long-form creative writing: their generated stories either fall far short of the requested length, or their quality significantly degrades as length increases, especially when compared to frontier models. We present POLARIS (Policy Optimization with LLM-as-a-judge rewards and Anchored-Reference Injection for Storywriting), a lower-compute GRPO recipe with two key ingredients: a frontier LLM judge with a structured Story Quality rubric as the online reward, and human-reference injection (HRI), where a teacher-forced human-written story serves as a high-reward anchor within each GRPO group. By applying our training recipe to Qwen3.5-9B, using a dataset of approximately 1.4K prompt-story pairs derived from 100 short-story anthologies and 4 A100 GPUs, we obtain POLARIS-9B. Across five benchmarks spanning in-distribution and out-of-distribution prompts and rubrics, POLARIS-9B is competitive with much larger open-weight models while following length instructions more closely. A blinded human evaluation confirms that POLARIS-9B is preferred to the base Qwen3.5-9B and on par with Qwen3.5-27B. Despite training only on stories up to 4k words, POLARIS-9B preserves quality on prompts requesting stories up to 3 times the training length, a regime where most open-weight models degrade substantially in quality, length adherence, or both. More broadly, our results suggest that length generalization is a meaningful stress test for creative-writing models and a useful lens for distinguishing otherwise close models. --- ### Optimal Transport Flow Matching by Design Authors: Shimon Malnick, Matan Rusanovsky, Ohad Fried, Shai Avidan Published: 2026-06-02 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.04092v1 Flow matching models learn to transport samples from a simple prior distribution to a complex data distribution. When prior-data pairs are coupled via optimal transport (OT), the learned trajectories are straight and non-crossing, enabling fast, even single-step, generation. However, computing the OT coupling in high dimensions is intractable, and existing methods attempt to solve the OT problem, at the cost of persistent bias or significant overhead. Rather than solving for the OT coupling, we reformulate the problem. Once the prior is treated as a design choice rather than a fixed input, the OT coupling between prior and data is no longer unique. Many priors admit an OT-optimal identity coupling to the data, leaving us free to choose one that is also tractable to sample. We identify low-frequency projection of natural images as such a choice. The identity coupling between data and its low-frequency representation is empirically OT-optimal, the prior is structured enough to be sampled by a lightweight model at inference, and the remaining flow-matching task reduces to synthesizing high-frequency detail. Interpolating the prior with Gaussian noise further improves generation quality while preserving the OT coupling. The approach requires no modifications to the flow model itself, and integrates naturally with latent-space models, classifier-free guidance, and one-step generation frameworks. Across all benchmarks, our method reduces trajectory curvature by more than $2\times$ compared to existing flow matching methods, yielding better generation quality in the few-step regime. --- ### Large Language Models Hack Rewards, and Society Authors: Wei Liu, Xinyi Mou, Hanqi Yan, Zhongyu Wei, Yulan He Published: 2026-06-02 Categories: cs.LG, cs.AI, cs.CL, cs.CR, cs.CY arXiv: https://arxiv.org/abs/2606.04075v3 Reinforcement learning (RL) has become a dominant post-training paradigm, enabling large language models (LLMs) to learn from rewards. We observe that societal regulations are structurally similar to reward functions. They define measurable outcomes, thresholds, and exceptions, while often leaving institutional intent only partially specified. We hypothesise that the RL training process may exploit these gaps and therefore ask whether models' well-known tendency to hack reward functions during RL can scale into a more consequential failure mode named societal hacking: discovering loopholes in the rules society runs on. To study this phenomenon, we introduce SocioHack, a sandbox of 72 societal environments, and find that within these environments, reward hacking naturally emerges and leads to regulatory loophole discovery. Models learn to hack the social rules and generate strategies that remain technically compliant while defeating regulatory intent, and current LLM safeguards provide only limited mitigation. Therefore, collecting in-the-wild feedback for model training requires greater caution, and we need a next-generation post-training paradigm for safely iterating LLMs in real society.= --- ### Large Language Models Hack Rewards, and Society Authors: Wei Liu, Xinyi Mou, Hanqi Yan, Zhongyu Wei, Yulan He Published: 2026-06-02 Categories: cs.LG, cs.AI, cs.CL, cs.CR, cs.CY arXiv: https://arxiv.org/abs/2606.04075v2 Reinforcement learning (RL) has become a dominant post-training paradigm, enabling large language models (LLMs) to learn from rewards. We observe that societal regulations are structurally similar to reward functions. They define measurable outcomes, thresholds, and exceptions, while often leaving institutional intent only partially specified. We hypothesise that the RL training process may exploit these gaps and therefore ask whether models' well-known tendency to hack reward functions during RL can scale into a more consequential failure mode named societal hacking: discovering loopholes in the rules society runs on. To study this phenomenon, we introduce SocioHack, a sandbox of 72 societal environments, and find that within these environments, reward hacking naturally emerges and leads to regulatory loophole discovery. Models learn to hack the social rules and generate strategies that remain technically compliant while defeating regulatory intent, and current LLM safeguards provide only limited mitigation. Therefore, collecting in-the-wild feedback for model training requires greater caution, and we need a next-generation post-training paradigm for safely iterating LLMs in real society.= ---