Materials Science
Ultra-high vacuum Raman platform for in situ characterization of graphene nanoribbons
Jeong Ha Hwang, Amogh Kinikar, Lukas Rotach et al. (+9 more)
Atomically precise
graphene nanoribbons (GNRs) exhibit tunable electronic and
magnetic properties governed by edge
topology and finite-size effects, which make them versatile platforms for next-generation electronic and spintronic applications. However, the unpaired pi-electrons responsible for their
magnetic properties simultaneously make them highly susceptible to chemical degradation under ambient conditions. This intrinsic reactivity poses a central experimental challenge: accessing vibrational and electronic signatures of air-sensitive GNRs during synthesis and under controlled environments without breaking
vacuum. Once the material has been exposed to air, standard characterization techniques would probe oxidized or chemically modified species rather than the pristine form. Here, we overcome this limitation by developing a home-built ultra-high
vacuum (UHV) Raman platform designed to preserve sample integrity by preventing air exposure and to enable in situ investigation of material properties. The portable Raman
vacuum suitcase (RVS) integrates temperature control and precise gas dosing, allowing direct monitoring of growth kinetics,
lattice dynamics, and reactive-edge responses under well-defined thermal and chemical environments. Using this platform, we monitor the on-surface synthesis of 7- and 9-atom-wide armchair GNRs (7- and 9-AGNRs), quantify the
evolution of 7-AGNR Raman modes over a wide temperature range (162-748 K), and resolve chemical changes upon controlled O2 exposure that are consistent with oxidation at the reactive zigzag sites. These results establish UHV Raman spectroscopy with the RVS as a route to accessing the intrinsic vibrational signatures of low-dimensional
quantum materials under controlled environments.
Read the full abstract ▾
All authors Jeong Ha Hwang, Amogh Kinikar, Lukas Rotach, Andres Ortega-Guerrero, Carlo A. Pignedoli, Klaus Muellen, Thorsten G. Englmann, Xinliang Feng, Pascal Ruffieux, Roman Fasel, Mickael L. Perrin, Gabriela Borin Barin
Data Structures and Algorithms
Competitive Random-Order Correlation k-Clustering
Mahsa Derakhshan, Andisheh Ghasemi, Rajmohan Rajaraman et al. (+2 more)
Correlation clustering has been extensively studied over the last two decades in many different computational models owing to its wide-ranging practical applications. The goal is to compute a partition of the vertex set such that the total number of disagreements, i.e. the sum of edges between clusters, and non-edges within clusters, is minimized. In this paper, we study correlation $k$-clustering, in which the total number of clusters is restricted to $k$. Correlation $k$-clustering is NP-hard, and while previous work has presented a
polynomial time approximation scheme for constant $k$, there are no known results for general $k$. Our first result is a
polynomial-time constant-factor approximation
algorithm for correlation $k$-clustering for general $k$. The main focus of this work is in the more challenging online setting. Noting that the best competitive ratio under
adversarial arrivals is known to be $\Omega(n)$, we concentrate on the well-studied random-order model, where vertices arrive in random order and on arrival of a vertex, edges to its earlier-arrived neighbors are revealed. We prove a surprising lower bound of $\Omega(\log k)$-competitiveness for any online
algorithm, which can be extended to $\Omega(\log n)$ when $k = \text{poly}(n)$. Finally, the main result of this paper is a polylogarithmic upper bound on the competitive ratio for correlation $k$-clustering, using an
algorithm inspired by the classic Pivot
algorithm for correlation clustering. All
theorems have been verified using Lean.
Read the full abstract ▾
All authors Mahsa Derakhshan, Andisheh Ghasemi, Rajmohan Rajaraman, Omer Wasim, Tegan Wilson
Multiagent Systems
Positional choice and robust collective behavior in fish schools: biohybrid experiments and modeling
Vahagn Grigoryan, Donato Romano, Cesare Stefanini et al. (+1 more)
Collective behavior of fish schools is usually modeled on the assumption that each individual follows specific rules of motion that depend on its position and velocity relative to its neighbors. Although these models reproduce many schooling patterns observed in nature, it remains unclear whether the assumed rules are realistic at the individual level. To address this question, we first analyzed a set of experiments in which a live fish interacted with four moving robotic fish in a tank, and measured the time it spent in each position relative to the
robots. We then simulated this experiment, replacing the fish with an
agent, and assessed the extent to which classical models agree with the experimental observations. An extensive exploration of the parameter space showed that these models closely matched the experimental observations, with a very specific choice of parameters. However, they were highly sensitive to the parameter values: a minor perturbation caused them to fail completely. To resolve this, we propose a new model that incorporates an additional exploratory term characterizing the
agent's positional preference at every moment. This model proved robust to small errors in the parameters while successfully replicating the experiments. Furthermore, when generalised to multiple fish, the model reproduced schooling behaviour and common schooling patterns, while also exhibiting exploratory behaviour at both the individual and the school level. This shows that social cohesion coexists with individual exploratory behaviour, and that realistic models of collective behaviour should account for both social interactions and this exploratory component.
Read the full abstract ▾
All authors Vahagn Grigoryan, Donato Romano, Cesare Stefanini, Giulia De Masi
Machine Learning
Justin Deschenaux, Alexandre Galashov, Andrew Campbell et al. (+4 more)
Diffusion models have revolutionized generative modeling for continuous data through the gradual refinement of a belief state. This iterative refinement has not yet carried over to discrete
diffusion models, which discard uncertainty at intermediate steps through categorical sampling (information collapse). We propose Simplex
Diffusion Models (SDMs), a framework that lifts the
diffusion process to the probability simplex to represent beliefs over categories. SDMs admit probability paths with closed-form reverse transitions and can be trained with a simple cross-
entropy loss. Contrary to earlier proposals such as Dirichlet Flow Matching which requires integrating an ordinary differential equation, we introduce a DDIM-like sampler with a tunable level of stochasticity. Because SDMs operate on samples on the simplex, they can carry uncertainty across denoising steps, which mitigates information collapse. On OpenWebText, SDMs are competitive with strong Discrete
Diffusion baselines, achieving $17.0$ GenPPL at $5.46$ unigram
entropy in 64 sampling steps, close to real validation data. Even without Self-Conditioning (SC), SDMs outperform masked and uniform
diffusion (with SC or predictor-corrector sampling) on code generation (TinyGSM, $T=0.1$; $49.0\%$ vs. $45.8\%$). Distilled down to 8 steps, SDMs solve $32.1\%$ of GSM8K problems, more than distilled Discrete
Diffusion models with 128 steps ($21.4\%$).
Read the full abstract ▾
All authors Justin Deschenaux, Alexandre Galashov, Andrew Campbell, Li Kevin Wenliang, James Thornton, Arnaud Doucet, Valentin De Bortoli
Cryptography and Security
INTCC: A Framework for Interactive Confidential Computing
Qingzhe Bing, Kaiyuan Zhang and Yinqian Zhang
Confidential computing leverages Trusted Execution Environments (TEEs) to ensure the confidentiality and integrity of data in use. However, TEEs rely on remote attestation to guarantee the integrity of their initial
memory state. This model is fundamentally at odds with interactive development workflows. In scenarios like LLM fine-tuning and exploratory data analysis, data processors need human-in-the-loop capabilities, including dynamic code injection, intermediate state inspection, and hyperparameter tuning, all of which inherently violate the static, one-time integrity guarantees of traditional remote attestation. To reconcile this tension, we propose the interactive confidential computing paradigm, a system architecture enabling untrusted data processors to execute dynamic, non-deterministic operations within TEEs without compromising data confidentiality. Driven by the insight that inherently unmeasurable human interaction must be excluded from the Trusted Computing Base (TCB), we logically partition the TEE into an interactive controller and a verifiable runtime. To realize this paradigm, we present INTCC, a framework featuring three key mechanisms: (1) a proxy-based dispatch system to preserve the native development experience; (2) a fine-grained information flow control mechanism based on a security
lattice to prevent data leakage; and (3) a
privacy-preserving verifiable execution mechanism to guarantee the runtime compliance of dynamic workflows. We implement INTCC on AMD SEV-SNP using Confidential Containers and evaluate it across diverse real-world workloads. Our experiments demonstrate that INTCC effectively balances security and interactivity, incurring a practical overhead of less than 5% for LLM fine-tuning and under 17% for data analysis relative to baseline execution.
Read the full abstract ▾
Artificial Intelligence
BaRe-Mem: Bayesian Reliability Memory for Robust and Adaptive Agent Consultation
Peilin Feng, Zhengyang Huang and Soujanya Poria
In multi-
agent systems, reliable consultation is challenging because advisor capabilities vary across tasks, and misleading information can make consultation worse than autonomous
reasoning. We introduce BaRe-Mem, an online
Bayesian reliability
memory for multi-
agent consultation. It estimates advisor reliability based on the central model's internal belief representations and updates these estimates from historical interactions. These estimates modulate the influence of advisor responses and guide the choice between consultation and autonomous
reasoning. Across nine
benchmarks and six central models, BaRe-Mem is more robust to misleading advisor information than debate and majority voting. On the more challenging tasks, it remains above autonomous
reasoning across all tested misleading levels. Moreover, we extend the BaRe-Mem mechanism to worker allocation in
agent teams. On the MuSiQue
benchmark, BaRe-Mem improves task completion over routing by historical success counts and identifies capable workers earlier.
Read the full abstract ▾
Combinatorics
Nonnegativity of the $g$-polynomial of split matroids
Alice L. L. Gao and Matthew H. Y. Xie
We prove that the $g$-
polynomial of every split matroid has nonnegative coefficients, establishing Speyer's
conjecture for a class closed under taking minors and containing all paving and copaving matroids. Our proof uses a deletion–contraction identity obtained by constructing an auxiliary split matroid. We first show that every simple, cosimple, connected split matroid $M$ has an element $e$ for which both the deletion $M\setminus e$ and the contraction $M/e$ are connected. More generally, let $M$ be any connected split matroid of rank $k$ on a ground set $E$ with $|E|\ge4$. Suppose that $e\in E$ is such that both $M\setminus e$ and $M/e$ are connected. For every $v\in E\setminus\{e\}$, we construct a connected elementary split matroid $N_{e,v}$ of rank $k-1$ on $E\setminus\{e,v\}$ satisfying \[ g_M(t)=g_{M\setminus e}(t)+g_{M/e}(t)+t\,g_{N_{e,v}}(t). \] Using a fixed total order on $E$, we prescribe the proper cyclic flats of $N_{e,v}$ and their ranks. The identity follows from the covaluative formula of Ferroni and Schröter together with recurrences for its correction
polynomials, derived from Ferroni's enumeration of admissible Delannoy paths for Schubert matroids. Since all three matroids on the right have fewer elements, the identity supplies the induction step in the proof of nonnegativity.
Read the full abstract ▾
Artificial Intelligence
RareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis
Bo Zhang, Yuchen Wang, Dongbai Li et al. (+6 more)
Rare-disease diagnosis is a long-tail
reasoning problem: phenotypes are incomplete, individual disorders are sparsely documented, and relevant evidence is distributed across ontologies, gene annotations, and biomedical text.
Language models consequently favor common conditions, miss rare candidates, or produce plausible but invalid names. We introduce RareDx, which couples controlled evidence use with knowledge-graph-grounded policy optimization. RareDx-Harness normalizes heterogeneous records into one ranked-diagnosis task and compares direct inference, static
retrieval, adaptive tools, and structured phenotype-gene-disease
reasoning over a shared knowledge layer. The training pipeline combines Top-10 post-training with RareDx-KGPO, our knowledge-graph-grounded policy optimization method. Its reward projects predictions into a canonical disease graph and integrates curated graded relevance, ontology proximity, biomedical similarity, and phenotype consistency. Vocabulary and output-budget constraints prevent dense partial credit from rewarding fabricated or overlong differentials. Across eight
benchmarks, the complete RareDx system centered on Qwen3.5-9B reaches 38.34 macro Hit@10, 1.60 points above GPT-5.5 under the archived protocol; a disjoint validation-selection audit retains a 6.80-point routing gain over Direct on held-out cases. The 27B system reaches 23.53/36.56/40.76 at Hit@1/5/10. Controlled ablations show that
retrieval is not uniformly helpful and that controlled routing is central to the gain. These results indicate that structured medical knowledge can turn a compact model into a competitive diagnostic ranker across heterogeneous long-tail settings in clinical practice.
Read the full abstract ▾
All authors Bo Zhang, Yuchen Wang, Dongbai Li, Matthew Yu Heng Wong, Qingkai Zeng, Lijun Wang, Tien-Yin Wong, Peng Cui, Tianyu Liu
Number Theory
A note on large clusters of $E_2$-numbers
Genheng Zhao
Let $q_n$ denote the $n$th product of two distinct
primes. By completing a square in Sono's sieve and solving the resulting one-dimensional variational problem, we prove unconditionally that, for every $\varepsilon>0$ and all sufficiently large integers $\rho$, \[ \liminf_{n\to\infty}(q_{n+\rho}-q_n) \leq \exp\left((\pi+\varepsilon)\sqrt\rho\right). \]
Read the full abstract ▾
Quantum Physics
Sublinear Copies Suffice for Fidelity Estimation with Pauli Measurements
Jayadev Acharya, Abhilash Dharmavarapu, Yuhan Liu et al. (+1 more)
We present the first worst-case, sublinear protocol to estimate the fidelity between a known target state and unknown lab-prepared state using Pauli basis measurements. More precisely, for $d$-dimensional states, we estimate the fidelity to $\pm \varepsilon$ using $o(d^{0.9675})/\varepsilon^2$ copies, which applies to all target states. Furthermore, our results establish improved guarantees for Pauli shadow
tomography and Distributed Inner Product Estimation.
Read the full abstract ▾
All authors Jayadev Acharya, Abhilash Dharmavarapu, Yuhan Liu, Nengkun Yu
High Energy Physics (Phenomenology)
Heavy Quark Recombination Across Collision Systems: Meson and Baryon production
Vincenzo Minissale, Salvatore Plumari, Maria Lucia Sambataro et al. (+2 more)
Measurements of heavy-
flavor baryons in
nucleus–
nucleus and
proton–
proton collisions have provided strong evidence that heavy-
quark hadronization cannot be described solely in terms of
vacuum fragmentation. In particular, the large enhancement of the $\Lambda_c/D^0$, $\Xi_c/D^0$ and $\Omega_c/D^0$ ratios observed at RHIC and LHC energies has been successfully described within a coalescence plus fragmentation approach, supporting the relevance of recombination mechanisms in both large and small collision systems. Building on these results, we discuss recent developments in the coalescence hadronization framework. Particular emphasis is given to the production of multi-charmed
baryons, namely $\Xi_{cc}$, and $\Omega_{ccc}$, whose yields are predicted in a wide range of collision systems from PbPb to KrKr, ArAr and OO collisions. The strong sensitivity of these states, especially $\Omega_{ccc}$, to the underlying charm-
quark distribution makes them valuable probes of charm thermalization and non-
equilibrium effects in the QGP. We also present recent extensions of the model to the beauty sector, including predictions for $\Lambda_b/B^0$, $\Xi_b/B^0$ and $\Omega_b/B^0$ production. The results indicate sizeable recombination effects and a stronger
baryon enhancement than in the charm sector, highlighting the increasing role of coalescence for heavier
quarks. Finally, we briefly comment on ongoing developments toward a unified description of open and hidden heavy
flavor through the application of the same framework to quarkonium production in small collision systems.
Read the full abstract ▾
All authors Vincenzo Minissale, Salvatore Plumari, Maria Lucia Sambataro, Gloria Calà, Vincenzo Greco
Machine Learning
Graph World Models for Constrained Epidemic Policy Planning
Yiqi Su, Rashed Shelim, Lingyi Wang et al. (+2 more)
Epidemic policy planning often requires coordination between geographical regions, taking into account mobility-driven spillovers and how to make use of limited resources. Existing methods either lack action-conditioned models of coupled dynamics or cannot guarantee per-period feasibility. We present EpiMind, a graph world model framework for constrained epidemic policy planning across regions. A graph-factored recurrent state-space model generates joint policy-conditioned rollouts from regional latent beliefs, while graph-temporal ADMM optimizes regional interventions, enforces shared-resource feasibility through projection, and evaluates temporal specifications under the learned model. EpiMind reduces admission RMSE by 29% relative to graph-free dynamics modeling, plans within 1-5% of the best feasible constant policy with guaranteed shared-budget feasibility, and outperforms all deployable baselines across three resource budgets in real-context evaluation. These results demonstrate that graph-structured policy imagination with explicit constrained coordination supports effective epidemic interventions from learned dynamics.
Read the full abstract ▾
All authors Yiqi Su, Rashed Shelim, Lingyi Wang, Walid Saad, Naren Ramakrishnan
Computation and Language
Less Sycophancy, Stronger Refusal? Lessons for AI Safety from Mechanistic Interpretability
Xu Wang, Difan Zou and Xuansheng Wu
Reliable refusal of harmful requests is essential to the safe deployment of
language models. Because excessive eagerness to please users may undermine existing refusal capabilities, reducing sycophancy offers a potential route to stronger refusal beyond the harmful scenarios covered by safety training. We investigate this possibility using compensatory feature injection (CFI), a training technique designed to limit the acquisition of a target concept by supplying its associated activation during learning. Across three Qwen3.5 base models, we use sparse autoencoders (SAEs) to identify the top-ranked sycophancy feature from paired sycophantic and independent responses, then validate its behavioral influence through inference steering. We subsequently inject the selected feature during supervised fine-tuning on sycophantic targets. Positive injection reduces learned sycophancy after removal (by 62.0% relative to ordinary fine-tuning in 35B-A3B), whereas modest negative injection increases it. Unexpectedly, these reductions in sycophancy do not consistently improve direct refusal of harmful requests, motivating a narrower evaluation of the same harmful intents under user pressure. In this setting, ordinary fine-tuning on sycophantic responses substantially weakens refusal, while selected checkpoints trained with positive injection recover part of the loss, including approximately 95% in 35B-A3B. These findings show that persistent sycophancy reduction does not guarantee stronger direct refusal, while identifying recovery under user pressure as a distinct, conditional benefit of training intervention.
Read the full abstract ▾
Representation Theory
Automorphic $C^*$-algebras of reductive groups
Jun Yang
Let $G$ be a reductive group over a number field $F$. We define $C^*_{\rm aut}(G(\mathbb A))$, the automorphic $C^*$-
algebra of $G$, to be the $C^*$-algebraic image of the full automorphic representation of $G(\mathbb A)$ on $L^2(G(F)\backslash G(\mathbb A))$. Using the Langlands spectral decomposition with respect to discrete Levi data, we construct an injective $*$-homomorphism \[ C^*_{\rm aut}(G(\mathbb A)) \longrightarrow \bigoplus_{[M,\sigma]} K_{C_0(\widehat{A_M})} \left( \operatorname{Ind}_P^G C_0(\widehat{A_M},H_{M,\sigma}) \right)^{W(G,M,\sigma)}, \] where $[M,\sigma]$ ranges over the associate classes of discrete Levi data. When $G$ is $\operatorname{GL}(n)$ or an inner form of it, we prove that this map is an isomorphism of $C^*$-
algebras.
Read the full abstract ▾
Group Theory
Classification of finite left-simple skew braces
Marco Damele
We give a complete classification of finite left-simple skew braces. We prove that a non-trivial finite skew brace has no non-trivial proper left ideals if and only if it is either the trivial skew brace of
prime order or the almost trivial skew brace associated with a finite non-abelian simple group.
Read the full abstract ▾
Machine Learning
Learning the Robustness Mechanism with Bilevel Optimization
Yiyang Shen, Qihang Lin and Weiran Wang
We propose a distributionally robust learning framework where parameters defining the robustness mechanism are learned from held-out data instead of extensively tuned. Using bilevel optimization with both upper and lower level minimax problems, we create two instances of our framework to tackle setups with and without group labels in the training set. Theoretically, we provide sample complexity analysis for our robustness mechanism learning paradigm, showing that it achieves generalization guarantees comparable to exhaustive grid search while being more computationally efficient. Empirically, we evaluate our framework under a challenging setup when both intra-group and inter-group test distribution shifts occur at the same time, thereby demonstrating the efficacy and scalability of our method.
Read the full abstract ▾
Artificial Intelligence
Continuous Context Management
William Hoy, Jingxuan Fan, Nurcin Celik et al. (+1 more)
Long-horizon large
language model (LLM)
agents commonly retain their complete interaction history until compaction is triggered at a predefined threshold. We study Continuous Context Management (CCM), which performs compaction at every turn to prevent interaction history from accumulating in the active
prompt. At each turn, a CCM
agent emits an updated
memory together with an environment action; its next
prompt contains the original task, retained
memory, and newest observation rather than the complete transcript. We first evaluate CCM without fine-tuning on TerminalBench-2 using Claude Sonnet 4.6, Claude Opus 4.6, GLM-5, and Kimi K3. CCM substantially reduces cumulative input usage and active-
prompt size, although it lowers task success for most models while preserving performance for Kimi K3. We use GRPO with privileged full-history
distillation to improve CCM in open-weight models. A frozen copy of the student's initial model scores each sampled student action under the complete history reconstructed from that student's rollout, providing dense action-
token supervision without a separate teacher rollout or reference solution. On WebShop, this objective substantially improves CCM over GRPO at both evaluated model scales and surpasses full-history GRPO for Qwen3-4B-Instruct, though not for Qwen3-8B. On Endless Terminals, the augmented method provides a modest improvement over GRPO, with both CCM policies outperforming the untrained full-history baseline. These results demonstrate that CCM is a viable inference paradigm for
agents operating with substantially reduced retained context and that its performance can be improved through reinforcement learning with privileged full-history
distillation.
Read the full abstract ▾
All authors William Hoy, Jingxuan Fan, Nurcin Celik, Xu Pan
Computer Vision
Learning to Reason with Persistent Object States for Video Instance Segmentation
Yongxue Xu, Boxue Yang, Ziqian Liu et al. (+3 more)
Video segmentation models maintain object identities by carrying instance information across frames. Under prolonged occlusion, reappearance, or interactions between similar instances, however, an unreliable update can overwrite a valid history and cause persistent identity drift. We introduce POSReasoner, a trainable, plug-and-play framework that explicitly decides when an observation should change an object's state. Each persistent state records identity, confidence, and absence history. A sparse state-observation graph supports Propose-Verify
reasoning: provisional associations are revisited using object history, predicted presence, and competition among identities. The verified decisions determine whether to retain, update, reactivate, or suppress each state, while a learned gate controls the evidence written back to
memory. Only verified transitions update the persistent state used in subsequent frames. POSReasoner uses standard
video annotations and keeps the base model frozen, enabling integration with diverse VOS and VIS architectures. Experiments across long-term VOS and VIS
benchmarks show consistent improvements over strong baselines, with the largest gains under occlusion and object reappearance.
Read the full abstract ▾
All authors Yongxue Xu, Boxue Yang, Ziqian Liu, Shaoqiu Zhang, Rui Qian, Haopeng Chen
Numerical Analysis
Finite Element Approximation of a Hemivariational Inequality for Steady-State Heat Conduction: Double-Limit Convergence of Penalization and Discretization
Piotr Bartman-Szwarc, Anna Ochal and Domingo A. Tarzia
In this paper we study the numerical approximation and convergence analysis of a steady-state heat conduction problem with mixed boundary conditions. The physical model is governed by a hemivariational inequality depending on a heat transfer parameter $\alpha > 0$. We consider the finite element approximation of this penalized problem, as well as its corresponding limit problem with a prescribed constant temperature on a part of the boundary. The main theoretical result establishes the strong convergence of the discrete solutions. Specifically, we prove the double-limit convergence of the finite element approximations to the limit solution as the mesh size $h$ tends to zero and the penalization parameter $\alpha$ tends to infinity, independently and simultaneously. We further complement the convergence analysis with an error estimate of optimal order for the discrete limit problem, and we explain why an estimate uniform in the penalization parameter cannot be expected. The theoretical results are illustrated by numerical simulations on four examples verified by the method of manufactured solutions.
Read the full abstract ▾
Machine Learning
Optimal Networks for Agentic Information Aggregation
MohammadHossein Bateni, Zahra Hadizadeh, MohammadTaghi Hajiaghayi et al. (+2 more)
We study information aggregation in the networked learning model introduced by Kearns, Roth, and Ryu (SODA 2026). There is a fixed distribution over $d$ features and a common label.
Agents learn in topological order on a directed acyclic graph. Each observes a subset of the features and its parents' predictions, fits a linear predictor to minimize mean squared error, and passes only its prediction forward. The global predictor is the best linear predictor using all features. Kearns, Roth, and Ryu show that the output
agent's error approaches the global predictor's error along sufficiently deep paths with suitable feature coverage, while insufficient depth can prevent aggregation even in large networks. In contrast to their main focus on a given graph and feature allocation, we consider the limits of the model under two settings. In the adaptive designer setting, a designer chooses the graph, feature allocation, and output
agent knowing the distribution. In the oblivious designer setting, the designer fixes all three before an adversary chooses the distribution. Each
agent observes one feature and receives predictions from a limited number of parents. We call the aggregation exact when the output
agent matches the global predictor exactly. For $d\ge3$, we show that no finite depth guarantees exact aggregation for every distribution with one parent per
agent, even when the designer knows the distribution. In contrast, two parents per
agent suffice for exact aggregation even in the oblivious designer setting. A fixed graph, feature allocation, and output
agent achieve this for every distribution at depth $O(d\log d)$. Knowing the distribution reduces the depth to $O(d)$. Both constructions use $O(d^2)$
agents, with a very large constant for two parents. We show the bounds on the depth and number of
agents are all optimal up to constant factors.
Read the full abstract ▾
All authors MohammadHossein Bateni, Zahra Hadizadeh, MohammadTaghi Hajiaghayi, Mahdi JafariRaviz, Shayan Taherijam
Computer Vision
Look Before You Judge: Training-Free Region Mining for Grounded and Explainable Deepfake Detection
Chia-Ling Chen, Yu-Ting Ta, Jian-Yu Jiang-Lin et al. (+8 more)
Multimodal large
language models (MLLMs) can explain deepfake verdicts in natural
language, but such explanations are not necessarily visually grounded in the visual evidence underlying the prediction. A model may describe plausible artifacts inferred from
language priors rather than from image evidence. Existing grounding methods improve visual reliance through decoding or
attention interventions, but they generally strengthen grounding over the entire image, making them ill-suited for forensic artifacts that are subtle, spatially localized, and image-dependent. We propose Look Before You Judge, a training-free framework that formulates explainable deepfake detection as a sequential evidence acquisition process. Instead of directly predicting image authenticity from holistic visual
reasoning, our framework first identifies image-specific candidate evidence regions by contrasting the MLLM's decoder-to-visual
attention between an original image and its Gaussian-blurred counterpart. The identified regions are then inspected individually, and the resulting local evidence is integrated with the global image context before reaching a final verdict. The framework operates without manipulation masks, external forensic models, or parameter updates, making it directly applicable to off-the-shelf MLLMs. Across five open-source MLLMs on TriDF and MMTD-Set, our framework improves detection accuracy by up to 12.8%, reduces CHAIR by up to 33.4% and
hallucination rate by up to 21.3%, and outperforms representative training-free decoding and
attention methods.
Read the full abstract ▾
All authors Chia-Ling Chen, Yu-Ting Ta, Jian-Yu Jiang-Lin, Tai-Ming Huang, Ling Lo, Po-Ching Chen, Yan-Tsung Wang, Pei-Heng Li, Ling Zou, Hong-Han Shuai, Wen-Huang Cheng
Quantum Physics
Analysis of asymmetric errors in NISQ experiments
Gil Kalai, Tomer Shoham and Carsten Voelkmann
This paper continues our analysis of NISQ experiments and especially the Google 2019 “
quantum supremacy" experiment. In arXiv:2008.05177 we proposed statistical tools to analyze symmetric and asymmetric readout errors, and the symmetric case was extensively developed and studied in arXiv:2404.00935, where we used
Fourier methods combined with statistical tools. Here we extend and complement our earlier study and analyze asymmetric readout errors for data from experiments and simulations. Our study was further motivated by a paper by Fefferman et al. arXiv:2306.16659, who showed that the effect of amplitude damping gate errors is to have larger fractions of zeroes in the samples produced by the
quantum computer. We show that this effect of gate errors is not observed in the Google experimental data, where, for the entire range of the number of
qubits, $12 \le n \le 53$, the fraction of zeroes and ones reflects only the asymmetry of reported readout errors, and report on simulations with $n=12,14$ using IBM's and Google's simulators.
Read the full abstract ▾
Signal Processing
Privacy-Aware ISAC for Full-Duplex Monostatic Systems Using Movable Antennas
Yasas Savinda, Mohammadali Mohammadi, Himal A. Suraweera et al. (+1 more)
This work investigates sensing
privacy in full-duplex (FD) monostatic integrated sensing and communication (
ISAC) systems with movable
antennas (MAs). The proposed approach jointly optimizes beamforming and
antenna trajectories to create a deceptive dummy DD-bin response at a passive sensing eavesdropper (Eve), while satisfying a true-bin sensing-quality requirement at the base station (BS). The resulting problem is highly non-convex. {To address this, a stage-wise alternating local-search framework is developed to obtain suboptimal solutions. Within this framework, we maximize the worst-case margin between dummy and true delay-Doppler (DD)-bin detector-oriented SINR surrogates over a discretized uncertainty region for Eve, incorporating detector-aligned dummy-bin refinement and true-bin preservation.} Simulation results show that the proposed MA-enabled design suppresses Eve's true-target DD-bin selection and increases dummy-bin selection probability compared with
benchmark schemes, while maintaining reliable BS sensing performance.
Read the full abstract ▾
All authors Yasas Savinda, Mohammadali Mohammadi, Himal A. Suraweera, Henk Wymeersch
Astrophysics of Galaxies
The abundance and emission of PAHs from the local to early Universe
Massimiliano Parente, Desika Narayanan, Paul Torrey et al. (+8 more)
We present the first large-volume cosmological simulation of
galaxy evolution with a model for the abundance and luminous emission of Polycyclic Aromatic Hydrocarbons (PAHs) across a wide range of
cosmic time ($0 < z \lesssim 6$). The simulation is built on a semi-analytic model (SAM) that includes a treatment of
dust size distribution, with PAHs being identified with the smallest carbonaceous grains ($< 15$ A) residing in the diffuse ISM of
galaxies. PAH emission is computed by post-processing the simulation output with the radiative transfer code GRASIL. In our model, where
stars only inject large grains, the PAH abundance emerges from the growth and shattering of
dust grains in the ISM, naturally reproducing the observed correlation between the PAH mass fraction ($q_{\rm PAH}$) and both
metallicity and specific
star formation rate (sSFR). We derive a relation to infer $q_{\rm PAH}$ from $L_{\rm PAH}$, $L_{\rm IR}$, and sSFR, with a relatively small scatter (a factor of $\approx 1.3$). Simulated PAH emission correlates well with key
galaxy properties such as SFR and molecular gas content, and reproduces observational constraints. Taken together, these results support the viability of a top-down formation scenario for PAHs. We extend our model predictions to higher
redshifts, which can be exploited for future missions such as PRIMA. This work is a significant step toward leveraging PAH observations as a tool for studying
galaxy evolution.
Read the full abstract ▾
All authors Massimiliano Parente, Desika Narayanan, Paul Torrey, Gian Luigi Granato, Brandon S. Hensley, Jed McKinney, Alexandra Pope, Gergö Popping, Irene Shivaei, Laura Silva, J.D.T. Smith
Artificial Intelligence
ARISE: Adapting to Evolving Capability Gaps in Agentic Reinforcement Learning
Kun Feng, Yuchen Fang, Yiyang Tan et al. (+6 more)
As a long-horizon
agent improves through experience, previously observed weaknesses may recede while new limitations emerge, continually changing what it still needs to learn. Yet the learning process often remains tied to a static view of these needs: fixed behavioral criteria and training priorities can become misaligned with evolving
agent capabilities, while sparse task-level feedback makes such misalignment more difficult to detect. Even when capability gaps are identified, rollouts from the current policy may repeatedly reproduce the same failures rather than explore better alternatives. To address this, we introduce Adaptive Rubric-Skill Co-
Evolution (ARISE), a reinforcement learning framework that uses rollout evidence to continually adapt evaluation criteria, exploration guidance, and training priorities. Rubrics evolve to reward partial behavioral progress, while their paired skills are refined and selectively activated to guide exploration toward unresolved weaknesses. Alongside this co-
evolution, capability-based adaptive sampling prioritizes tasks that target behaviors needing further improvement. Experiments on two challenging long-horizon
agent benchmarks, SkillsBench and Terminal-Bench, demonstrate that ARISE successfully enhances both overall task performance and training efficiency. The project page is at https://foundation-model-research.github.io/ARISE .
Read the full abstract ▾
All authors Kun Feng, Yuchen Fang, Yiyang Tan, Shuqi Gu, Yongxiang Zhao, Yu Liu, Xingyu Lu, Lintao Ma, Kan Ren
Computational Engineering
Latent Twin Operator
Deepanshu Verma, Riley Chen and Matthias Chung
Surrogate models deployed on real physical systems rarely see data at fixed resolutions: sensor configurations vary across deployments and may evolve over time as sensing infrastructure changes. We introduce the Latent Twin Operator (LTO), a latent-space surrogate for time-evolving PDEs with a Convolutional Conditional Neural Process (ConvCNP)-style encoder and decoder that accepts a context set of $N$ sensor observations with arbitrary placement and can be queried at any resolution. A learned latent
evolution map advances the encoded state directly between arbitrary time points, decoupling temporal
evolution from the observation and query discretizations. We derive an explicit $\mathcal{O}(N^{-2/(3D)})$ rate for the context discretization error in spatial dimension $D$ under quasi-uniform refinement. We verify the predicted decay empirically on a 2D heat-equation
benchmark. Across time-dependent PDE
benchmarks, LTO achieves strong accuracy under one-step comparisons and transfers from native to coarser spatial resolutions with fixed parameters. On Navier–Stokes, its direct latent
evolution further reduces error over longer prediction horizons relative to recursive evaluation.
Read the full abstract ▾
Computer Vision
AutoRef: Harness Optimization for Agentic Multi-Reference Image Generation
Yuta Oshima, Ku Onoda, Yusuke Iwasawa et al. (+3 more)
Recent image generation models can take multiple reference images as input and combine them into a new image. However, multi-reference image generation remains challenging: models may omit or duplicate subjects from the references, or produce images in which multiple subjects appear unnaturally pasted. Recent work has proposed image generation
agents that combine image generation models,
reasoning models, and a harness, which is an executable program that specifies how reference images are interpreted, how generation is performed, how outputs are diagnosed, and how the final image is selected. In multi-reference generation, however, references play different roles and outputs must satisfy many criteria at once, such as fidelity to each reference and the naturalness of the whole image, so many parts of the harness could be improved, from how references are processed to how outputs are diagnosed. This makes it hard to predict which changes will improve performance and by how much, and good harnesses difficult to design by hand; indeed, human-written harnesses vary widely in performance. We therefore propose AutoRef, which optimizes the harness automatically while keeping both models frozen: a coding
agent iteratively rewrites the harness code. AutoRef separates the tasks whose feedback informs proposals from the tasks used to select candidates, and continues the search from a beam of the top-ranked harnesses on the selection tasks. Using this procedure, we discover AutoRef-Harness, which improves the open-weight FLUX.2 [klein] 4B from 5.72 to 7.37 on held-out four-reference tasks of the MultiBanana
benchmark, matching or exceeding proprietary models including Nano Banana Pro and GPT-Image-1.5. Without re-optimization, the same harness also improves results when the generator, number of references,
benchmark, evaluator, or
reasoning model differs from those used in the search.
Read the full abstract ▾
All authors Yuta Oshima, Ku Onoda, Yusuke Iwasawa, Masahiro Suzuki, Yutaka Matsuo, Hiroki Furuta
Fluid Dynamics
Conservative local grid refinement for the vectorial lattice Boltzmann method
S.A. Hosseini, R. Strassle and I.V. Karlin
We present a conservative local grid-refinement coupling for the vectorial
lattice Boltzmann method (VLBM), a kinetic-type scheme for
hyperbolic conservation laws in which each conserved field is carried by its own population rather than by moment truncation of a Maxwellian. Coarse and fine regions share a common link speed and are advanced with different time steps, sub-cycled at a fixed ratio; a mean-preserving reconstruction in time supplies the missing coarse-to-fine boundary data, and a matched, Berger–Colella-style reflux returns the fine-to-coarse data, together conserving mass, momentum, and energy exactly, independent of the equation of state, the local relaxation parameter, or the time step. In addition, we propose an adaptive mesh refinement extension through a conservative pair of splitting and merging operators that require no modification to the underlying coupling. The approach is validated on a sequence of increasingly demanding two-dimensional cases, in every case recovering exact conservation and the expected accuracy gains over an equivalent uniform grid.
Read the full abstract ▾
Machine Learning
Let the Neurons Die: Exploiting ReLU-Induced Model Degradation
Kexin Li, Wenjun Qiu, Joshua Abraham et al. (+2 more)
Rectified linear unit (ReLU) networks can suffer from dying
neurons, where units with persistently negative pre-activations produce zero outputs, blocking gradients through their activations. To exploit this failure mode, we present three training-time availability attacks based on data ordering and poisoning. We begin with the basic dynamic data-ordering attack (DOA), which greedily constructs a training prefix by selecting the next example that minimizes the target layer's post-update weight sum, aiming to push ReLU units toward negative pre-activations without modifying training samples or labels. We then develop two poisoning attacks, IG-DOA and IG-SKA, which use gradient inversion to synthesize class-conditioned samples by matching reference gradients in adverse model states constructed through data ordering or soft knockout, respectively. Soft knockout rearranges weights across adjacent layers to concentrate negative contributions. On a fully connected ReLU network trained on MNIST, ordering 100 of 60,000 training examples reduces test accuracy from 96% to 95% after only five epochs. Adding 200 poisoned samples from a single class reduces test accuracy to approximately 86-88% after five epochs in most evaluated conditions, compared with approximately 96% under clean training. These results demonstrate that ReLU-targeted data ordering and poisoning can impair learning without directly modifying the victim model's parameters.
Read the full abstract ▾
All authors Kexin Li, Wenjun Qiu, Joshua Abraham, Aditi Maheshwari, David Lie
Machine Learning
GeoGAE: Scalable Graph-Level Autoencoding via Hyperball Cloud Representations
Radosław Nowak, Anna Bielawska, Bogusz Stefańczyk et al. (+2 more)
Embedding structured objects into Euclidean spaces has enabled a wide range of successful machine learning applications. Such objects include words, documents, image patches, time series, and graph nodes. In contrast,
embedding entire graphs remains a challenging problem. Existing methods either sustain the original order of the graph nodes or match the output nodes to the input ones, both of which create scalability issues. In this work, we propose a graph representation as a cloud of hyperballs, which allows us to define a specific, typically unique, node ordering. Based on this representation, we propose GeoGAE, an autoencoder, in which the
Transformer encoder translates a hyperball cloud into a graph-level
embedding, and the
Transformer decoder translates the graph-level
embedding back into the graph. This formulation enables the model to capture both the global graph structure and local relational patterns. We evaluate our method on multiple graph
datasets, spanning various domains. The results demonstrate effectiveness of our method in encoding and reconstructing graphs from their
embeddings.
Read the full abstract ▾
All authors Radosław Nowak, Anna Bielawska, Bogusz Stefańczyk, Maciej Sanocki, Paweł Wawrzyński