Peer Revenue Equity in science

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Sunday 30 August

Computation and Language

Cross-lingual Functional Vectors for Emotion Detection in Large Language Models

Function vectors (FVs) have recently emerged as a promising mechanism for steering the behavior of large language models (LLMs) by injecting task-specific latent direction representations derived from in-context demonstrations. While prior studies have shown that FVs can recover task behavior in structured in-context learning settings, their effectiveness on semantically complex tasks and their ability to generalize across languages remain underexplored. We investigate the cross-lingual transferability of FVs using multilingual multi-label emotion recognition as a challenging semantic classification benchmark. Specifically, we examine whether FVs extracted from a source language can steer task behavior in another language under both standard clean and perturbed zero-shot settings without providing demonstrations during inference. Across diverse cross-lingual settings, applying FVs substantially improves performance, suggesting that FVs capture language-agnostic, task-relevant signals rather than purely language-specific lexical patterns, and highlighting their potential as a lightweight and transferable mechanism for multilingual task adaptation. We observe that each LLM exhibits a relatively stable optimal range of attention heads for constructing effective FVs, and the pattern remains consistent across languages. In addition, FVs can partially replicate the task-steering effects of standard few-shot in-context learning while avoiding the computational overhead of processing multiple demonstrations, making them effective for large-scale practical applications. Our code is available at https://github.com/yingjie7/cross_lingual_fvs.
Read the full abstract

All authors Jieying Xue, Phuong Minh Nguyen, Minh Le Nguyen, Shogo Okada

Artificial Intelligence

LLMs Interpret, Embeddings Organize, Graphs Emerge: Agent-Driven Compilation of Scientific Knowledge

Sustained scientific work requires a knowledge substrate that carries interpretation across tasks and preserves paths to source evidence. We call this process scientific knowledge compilation and implement it in ASKS, the Agent-Driven Scientific Knowledge System. For each source, an LLM produces a readable Wiki view and machine-facing semantics. Deterministic checks convert the latter into a document-local GraphDelta, and embedding geometry together with explicit graph rules integrates the proposed changes into persistent state. Each ingest is an inspectable state transition over accumulated knowledge, with compiled Wiki and graph views linked to the preserved source record. We examine this process by chronologically compiling 56 published papers from one research program. Branch survival, cross-paper support, lineage, coverage, and churn yield a source-traceable author research portrait centered on tensor-network methods, with branches into quantum many-body research, tensor-network machine learning, and quantum-AI-oriented directions. In this run, higher-level Hub organization remains stable and low-churn. Canonical-node growth is predominantly additive. Graph-level measurements and navigation paths retain links to the source records from which they were compiled.
Read the full abstract

All authors Shi-Ju Ran, Kun Zhang, Xi Wu, Liu-Si Yang, Wen-Jun Li

Computer Vision

See the Change, Keep the Flow: Unsupervised Action Segmentation via Spectral-Temporal Representation Learning

Unsupervised action segmentation aims to discover latent action categories and their temporal organization without action annotations. Optimal transport-based methods provide structured frame-to-action assignments, however, their pseudo-label quality is fundamentally conditioned on the representation space used to construct the transport cost. We argue that reliable OT pseudo-labeling requires a representation geometry that is simultaneously sensitive to discriminative action changes and coherent along local temporal progressions. Based on this insight, we propose SpecT-OT, a spectral-temporal representation learning framework built upon an unbalanced optimal transport pseudo-labeling concept. SpecT-OT introduces a Spectral Reparameterization Projector (SRP), which parameterizes projector weights with fixed Fourier bases and learnable coefficients to improve the modeling of rapidly varying discriminative features, and Temporal Affinity Regularization (TAR), which imposes distance-aware, label-free constraints on pairwise frame affinities to stabilize local temporal structure. The two components jointly produce more discriminative and temporally stable transport costs, yielding more reliable pseudo-labels for iterative representation learning. Experiments on four benchmarks demonstrate strong performance compared with state-of-the-art methods. SpecT-OT achieves the best results on 13 of 15 metrics, including 4.1-point MoF and 7.4-point F1 gains over the baseline on Breakfast and Desktop Assembly, respectively.
Read the full abstract

All authors Yun Li, Jun Xiao, Cong Zhang, Kin-Man Lam

Computation and Language

Beyond Surface Alignment: Grounding the Dynamics of Situational Understanding and Generative Control in LLMs

The current alignment tuning paradigm for Large Language Models (LLMs) prioritizes surface-level behaviors – fluency, safety, and tonal consistency. While effective for casual chat, this thesis argues that such surface alignment masks a lack of grounding, creating models that are stylistically confident but situationally brittle. We propose a framework of Grounded Alignment, analyzing how models process context (Input) and structure generation (Output), then aligning these grounded behaviors to human needs. First, we evaluate failures in Situational Grounding. SitTest shows that despite large context windows, state-of-the-art models struggle to maintain a consistent "mental model" of a changing environment. ReCode further shows that models rely on surface heuristics rather than deep syntactic dependencies: they "read" extensive histories without truly "understanding" the evolving situation. Second, we evaluate Generative Grounding. We introduce the Branching Factor (BF) to map LLM generation, finding that standard alignment tuning constricts this landscape into premature stylistic collapse. Hindsight further shows that models often fail to understand their own generations. Finally, we propose Dynamic Control for grounded interaction. AI Realtor demonstrates context engineering to compensate for poor situational grounding. Base-Aligned Model Collaboration decouples exploration from stylistic constraints. We also present Annealed Sampling for verifiable reinforcement learning and apply these ideas to Addiction Support, where model-generated rationalization offers a communication interface for high-stakes domains. Collectively, this work moves beyond surface alignment toward agents anchored in both context and generation.
Read the full abstract

Computer Vision

nnMNet: Baseline for Martian Terrain Semantic Segmentation

Semantic segmentation is a crucial task for understanding Mars, the most Earth-like planet in our solar system. However, it is challenging because the Martian surface is highly unstructured and complex, making accurate pixel-level prediction and fine-grained annotation difficult. Recent advancements in deep learning have introduced numerous methods and datasets to address these challenges. Nevertheless, the field lacks a robust, publicly available, and reproducible baseline, as well as a unified benchmark to facilitate fair evaluations. In this work, we present nnMNet, a new baseline model designed for Martian terrain semantic segmentation. Building upon nnWNet, we integrate linear attention to better capture global context and employ lightweight convolutions to reduce computational overhead. To bridge the gap between local and global representations, we introduce the Spatially-Aware Fusion Block (SAFB), which augments and combines features with diverse characteristics. Furthermore, we establish a new benchmark by curating and standardizing three high-quality datasets for thorough evaluation. nnMNet achieves new state-of-the-art 86.61%, 83.25%, and 88.24% mIoU on SynMars-TW, SynMars-Air, and MarsScapes, respectively. Our code, models, and datasets are publicly available at https://github.com/dereklee0310/nnMNet.
Read the full abstract

Machine Learning

Wide Learning: Learning to Reach Evidence

Machine learning is usually evaluated after an evidence interface has been fixed. A dataset, sensor suite, query language, action set, or experimental protocol determines which observations can be obtained, and learning is judged by what it extracts from them. We study a complementary capability. A learner's state can determine which evidence-generating experiments it can reliably realise under bounded resources, even when primitive affordances remain fixed. We call this learner-relative experiment family its effective epistemic reach, and use Wide Learning for task-relevant learning-induced changes in that family.We formalise effective reach relative to learner state, deployment budget, reliability threshold, and evaluation distribution. In a controlled construction, two hidden worlds have exactly the same public observation law. An informative diagnostic exists in a fixed five-primitive substrate. Before calibration, one address attempt realises it with probability at most $2^{-10} = 1/1024$, below a pre-specified 0.95 threshold; after calibration, held-out realisation is 1. Public-channel total variation is 0, whereas the realised diagnostic has total variation 1, and sealed binary risk moves from approximately 1/2 to 0. The construction establishes that learning can change effective epistemic reach even when primitive affordances and deployment resources are held fixed. It opens a complementary evaluation question for learning systems: not only what they infer from available evidence, but what informative evidence experience teaches them to bring within reach.
Read the full abstract

Computer Vision

SnapBench: Benchmarking Snap-and-Ask Multimodal Retrieval for Mobile Interactions

Mobile AI acts as a visual oracle, empowering users to snap a picture of something and ask for information. Snap-and-ask retrieval is now one of the most common entry points for mobile AI, yet photos are often blurry, while text questions may be short or mistyped. Existing benchmarks only test on clean inputs or do not isolate paired robustness in snap-and-ask retrieval. Therefore, we introduce SnapBench, the first paired benchmark for robust snap-and-ask multimodal retrieval, spanning 1,145 queries, 9,085 gallery items under 53 controlled corruption conditions with human annotations. We evaluate 16 multimodal retrievers, covering dual-tower encoders and embedding-based VLMs. Results show that image corruptions substantially degrade retrieval, while text corruptions mainly affect text-only retrieval and have limited impact on joint retrieval. Clean image-only retrieval often outperforms joint retrieval, indicating the coarse-text drag and the lack of cross-modal fallback under noisy inputs. SnapBench provides a controlled testbed for evaluating robust retrieval in snap-and-ask scenarios. We further propose MOOR (Modality-anchored, Outlier-aware, Optimal Reweighting), a simple adaptive fusion approach, highlighting the need for reliability-aware modality calibration in snap-and-ask retrieval.
Read the full abstract

All authors Zirong Chen, Fuda Ye, Kuan Zhang, Enjun Du, Junfu Pu, Xinlei Wang, Xinyu Zuo, Lisheng Duan, Jin Ma, Yongqi Zhang

Computation and Language

Agent Zero Memory: Provenance-Aware Long-Term Memory for LLM Agents

Large language model (LLM) agents need durable, faithful memory of everything a user or organization has said and stored, yet most memory systems commit to a single organizing structure (a fact store, a vector index, or a knowledge graph) and inherit its blind spots. We present Agent Zero Memory, a provenance-aware long-term memory system that distils a user's conversations, files, and connected sources into three parallel memory systems, each capturing a different facet of the same history: an episodic Memory Events timeline that makes when and what changed first-class, an associative entity-event knowledge graph that links people and projects across sessions, and a semantic, curated, citation-locked Hierarchical Documentary Memory (HDM) of durable facts. A retrieval turn runs an intent gate (so self-contained turns add no latency), a source router, and three concurrent agentic searches, one per system, each a tool-using loop over hybrid (embedding + lexical) search under agent-controlled filters; their grounded, cited answers are integrated into one answer with a single confidence. We formalize the reading discipline: every learned item is a provenanced item carrying its origin, timestamp, and evidence pointer, and every answer is read under a citation lock, so it may cite only evidence its reader actually opened; fabrication is structurally excluded and the system abstains rather than guesses. On two public benchmarks the system sets a new state of the art: 95.60% on LongMemEval and 93.60% on LoCoMo, improving over the strongest prior systems by +0.73 and +1.10 points. A controlled study across eight backbone LLMs characterizes the accuracy-cost-latency frontier: accuracy varies by only 3.4 points while per-query cost varies by ~30x, with near-state-of-the-art quality at up to 20x lower cost per query, the signature of memory-driven, rather than model-driven, quality.
Read the full abstract

Computation and Language

Hindsight Memory-PRM: Supervising Memory Management with Auditable Hindsight Credit

Memory operations of long-horizon LLM agents are hard to supervise: an operation's value is unobservable when it is taken. But they are special – they leave machine-readable evidence in the trajectory: retrieval hits and answer-time citations. Hindsight Memory-PRM exploits this audit trail twice: offline to train an operation-conditioned memory-utility critic, and online, where retrievals, citations, and one controlled deletion-and-reanswer per probe settle an intervention-calibrated entry-level presence credit, propagated along version chains as an action-level proxy reward – no per-operation human labels, no Monte-Carlo replay of continuations. On held-out LoCoMo a local 8B policy reaches 77.5% under a fixed shared reader, surpassing its API teacher (65.1%) and all reproduced external systems, at one eighth the context of Mem0's official operating point; on LongMemEval, 79.0%. Ablations attribute the gain to causal calibration rather than signal density, and the policy converges to a multi-version memory organization whose gains no tested open-loop baseline reproduces.
Read the full abstract

All authors Haoxuan Jia, Yang Liu, Yingguang Yang, Yancheng Chen, Chongyang Zhang, Hao Zheng, Qian Li, Yulin Huang, Jianshen Zhang, Yongzhi Qi, Shang Luo, Kefu Xu, Hao Peng, Junyu Lu, Du Cheng, Philip S. Yu, Bin Chong

Information Retrieval

RePair: Turning Retrieval Failures into Counterfactual Hard Pairs

Vision-language retrieval with CLIP-style dual encoders achieves strong cross-modal performance, yet practical accuracy often hinges on localized semantic distinctions where top-ranked near misses differ from the true match by a single critical detail. Hard-sample mining can select confusable candidates but cannot construct corrected counterparts; synthetic augmentation can generate novel samples but, without conditioning on actual model failures, targets irrelevant dimensions of hardness. We observe that a top-ranked false positive is a counterfactual scaffold—sharing most of the query's semantics while differing in a localized failure-causing residual. Minimally correcting this residual yields a hard positive of the ground truth in the same modality; the corrected and unedited versions form a hard negative pair that straddles the decision boundary, producing complementary pull–push supervision. We introduce RePair, guided by three principles—Validity, Minimality, and Locality—which mines false positives bidirectionally, applies LLM-guided counterfactual editing, and trains with a local hard-pair contrastive objective. On Flickr30K and COCO30K, RePair outperforms controlled augmentation baselines with only 107K synthetic samples—26%–75% fewer than comparable methods—confirming failure-conditioned repair is more data-efficient than error-agnostic augmentation.
Read the full abstract

All authors Siyi Liu, Xiaorong Zhu, Enjun Du, Xinyu Zuo, Lisheng Duan, Haijin Liang, Jin Ma, Junfu Pu, Yongqi Zhang

Data Structures and Algorithms

FirstFit online coloring in the random order model

The average performance of FirstFit online coloring on trees in the random order model is completely determined in recent works of Frei et al. and Bosek et al., showing $\Theta(\log n /\log\log n)$ number of colors, improving the $\Theta(\log n)$ colors in the adversarial model. We provide a few further results on slightly more general graph classes. Firstly, we extend their method to obtain a simple path-counting principle for sparse graph classes, which immediately yields for example that cactus graphs and uniform hypertrees exhibit a similar improvement. We then show that FirstFit uses only $O(1)$ colors on crown graphs, a standard example where adversarial arrival forces $\Theta(n)$ colors. We further show that density alone (even linear minimum degree) is insufficient to guarantee $O(1)$ colors even on bipartite graphs. Finally, we identify graph classes, including unit interval graphs and some graphs of high chromatic number, for which random arrival provides only limited improvement. We end with some open problems.
Read the full abstract

All authors Xinyu Ye, Yuechuan Xu, Zixuan Wang, Jiaying Zheng, Yaqiao Li

Quantum Physics

Evolution Generators for Complex Parameters

Theoretical studies on how quantum systems are affected by external factors are often formulated through parameter changes in the system's Hamiltonian. Beyond Berry connections and phases, which focus on specific Hamiltonian eigenstates, recent studies inspired by non-Hermitian quantum formalisms provide a framework for obtaining general state evolution generators for real-valued parameters. This study extends the applicability of the evolution generator formalism to complex-valued parameters via Wirtinger derivatives. By treating a complex parameter and its conjugate as independent variables, the evolution equations are derived for both quantum states and the metric of the Hilbert space bundle. The analysis demonstrates that while state evolution with respect to a complex parameter is naturally governed by its corresponding evolution generator, metric evolution requires a coupled contribution from both the generator and its conjugate counterpart to preserve state normalization. Explicit examples are worked out to illustrate the implementation and physical consistency of the formalism.
Read the full abstract

Robotics

$N_0$-Foundation: Towards the Age of Tactile Intelligence

We present $N_0$-Foundation, a paradigm for tactile-enabled embodied manipulation, which integrates tactile sensing hardware, large-scale multimodal data, tactile representation learning, and standardized evaluation. First, we engineer the infrastructure for scalable data collection, including a vision-based tactile sensor, a tactile Universal Manipulation Interface (UMI), and a synchronized visuo-tactile data collection system supporting both robot embodiments and UMI-based demonstrations. Leveraging this infrastructure, we construct NeoData, which contains more than 30000 hours of synchronized visual and tactile demonstrations, spanning six embodiments, 450 tasks, and billions of paired RGB and tactile frames collected through a mixture of real-robot teleoperation and UMI-based demonstrations. To facilitate open research, we further release OpenNeoData, a 5000-hour open-source subset of NeoData. The dataset addresses a central limitation of existing manipulation corpora, critical for deformable-object manipulation, precise assembly, delicate force control, and sustained surface interaction. Capitalizing on the large-scale, heterogeneous tactile measurements, we propose NeoForce, a visuo-tactile representation model that learn transferable tactile representations across different sensor designs. To enable systematic evaluation of tactile embodied models built upon our infrastructure, datasets and tactile representations, we further propose a comprehensive benchmark, which combines the real-world NeoReal suite and the simulated NeoSim suite for standardized evaluation. Experiments across both suites show that policies benefit from the physical contact state rather than from the device-specific appearance of the tactile signal. We release the dataset, the representation, and the benchmark, aiming at supporting future work on tactile-enabled embodied manipulation.
Read the full abstract

Machine Learning

Adaptive Doubly Robust Off-Policy Evaluation for Ranking Policies under Diverse User Behavior

Off-policy evaluation (OPE) of ranking policies is challenging be- cause selecting and ordering multiple items from a candidate set makes the number of possible rankings grow combinatorially with the number of candidates and the ranking length. Consequently, Inverse Propensity Scoring (IPS), whose importance weight is the full-ranking probability ratio under the evaluation and logging policies, can have excessive variance. Independent IPS (IIPS) and Reward Interaction IPS (RIPS) reduce variance by imposing fixed assumptions on how users browse rankings, but may introduce bias when those assumptions mismatch actual behavior. Adaptive Inverse Propensity Scoring (AIPS) addresses this trade-off by adap- tively marginalizing importance weights over the actions that affect each position-wise reward. It attains minimum variance within a class of unbiased IPS-based estimators when the true user be- havior model is observed. However, its estimation accuracy may still degrade for longer rankings, and AIPS does not use a reward model for residual correction. We propose Adaptive Doubly Robust (ADR), which combines adaptive importance weighting with re- ward regression through a control-variate correction. We establish its unbiasedness when the true user behavior model is observed and characterize a sufficient condition under which it reduces vari- ance relative to AIPS. Across synthetic experiments with 10,000 simulations per condition, ADR improves mean squared error over AIPS and conventional ranking OPE estimators across a range of logged-data sizes and ranking lengths.
Read the full abstract

Dynamical Systems

Poisson actions of noncompact locally compact sofic groups have completely positive sofic entropy

We prove that every Poisson action of positive intensity of a noncompact locally compact second countable sofic group has completely positive measure sofic entropy with respect to every locally compact sofic approximation. The proof constructs random microstates for arbitrary measurable factors by completing local Poisson configurations with an independent exterior Poisson process. Probability estimates and a packing argument then show that every nontrivial factor has positive entropy. We also give a compact-group example showing that the noncompactness assumption is essential.
Read the full abstract

Machine Learning

On the Resilience of Text-to-Video Diffusion Models to Hardware Faults

We present the first systematic study of the resilience of text-to-video (T2V) diffusion models under random hardware-level faults. While T2V models are widely used for automated video generation due to their ability to produce high-quality, temporally coherent, and realistic videos, their iterative denoising process and spatiotemporal dependencies introduce unique failure modes. We perform an extensive fault-injection study covering both computational and memory faults across three T2V models and a representative benchmark. Our results show that (1) a single fault can degrade overall performance by up to 3.7%, with semantic correctness more affected than perceptual quality; (2) memory faults are more damaging than computational faults, high-order exponent bits are particularly vulnerable, and the widely-used bfloat16 is more susceptible than alternative formats; and (3) 7-28% of faults cause visible artifacts, including semantic changes such as added objects, suggesting that single faults are sufficient to alter output semantics. Our findings reveal reliability risks in deployed T2V systems and motivate further research on improving fault resilience. Code: \href{https://github.com/ztcoalson/T2V-Resilience}{https://github.com/ztcoalson/T2V-Resilience}.
Read the full abstract

All authors Zachary Coalson, A M Aahad, Stella Doehring, Zane Ma, Sanghyun Hong

Theoretical Economics

Credibility in school choice

In centralized school choice, a designer who announces a mechanism may deviate from it to favor some students without being detected. A mechanism is credible if it admits no such deviation. We study this credibility problem when students do not know others' reports and may observe only part of the assignment. Credibility is demanding: among common school choice mechanisms, only deferred acceptance is credible, and only when students observe enough of the assignment. We therefore rank mechanisms by their credibility. Deferred acceptance is more credible than any other stable mechanism and strictly more credible than the Boston mechanism, regardless of how much of the assignment students observe. Its ranking relative to top trading cycles and efficiency-adjusted deferred acceptance depends on disclosure: deferred acceptance is strictly more credible when students observe the entire assignment, whereas the ranking reverses when students observe only their own assignment.
Read the full abstract

Artificial Intelligence

A Systematic Survey of Agentic Skills: Architecture, Lifecycle, and Security

Autonomous large language model (LLM) agents increasingly face reliability, context consumption, and execution stability bottlenecks when deployed on complex, long-horizon tasks. While monolithic prompt engineering and stateless tool-calling paradigms struggle to scale, the field is rapidly converging toward agentic skills: modular procedural abstractions that externalize execution knowledge into reusable, executable, and portable artifacts. This paper establishes a unified systems foundation and reference architecture for the agentic skills ecosystem. We formalize skills as externalized procedural knowledge bridging high-level cognitive planning with deterministic execution environments, and systematically delineate the architecture across a nine-stage lifecycle: autonomous discovery, authoring and representation formats, memory storage, dynamic retrieval and routing, composition and orchestration, execution and repair, lifelong adaptation, empirical evaluation, and security governance. We further examine marketplace dynamics, public registries, and emerging adversarial threat vectors, alongside runtime verification and defense mechanisms. Finally, we categorize system implementations across software engineering, operating system navigation, embodied robotics, and scientific discovery, while highlighting critical open challenges in continual learning and benchmark realism. This work establishes agentic skills as a foundational paradigm for building scalable, robust, and verifiable autonomous language agents.
Read the full abstract

All authors Sanket Badhe, Deep Shah, Priyanka Tiwari, Nehal Kathrotia

Combinatorics

GPT, the Counterexample Machine

This document reports over 15 counterexamples found using GPT Pro over the course of 12 months, for problems in combinatorics, number theory, convexity, analysis, and other areas. The counterexamples concern both (reasonably) well-known problems as well as lesser known instances. The counterexamples are accompanied by the repo https://github.com/suvrit/count-ex-machina, and we warmly welcome new contributions toward helping building a rich “open-source countex library.”
Read the full abstract

High Energy Physics (Theory)

Higher-Spin dS/CFT beyond minimal type-A, and the doubled nature of de Sitter space

We study the holographic duality between free vector models and Higher-Spin Gravity in 4-dimensional de Sitter space. On both the bulk and boundary sides, we introduce new reality conditions, which allow us to include odd spins in the bulk, and to extend the duality from type-A to type-B (i.e. from boundary scalars to spinors). The new reality conditions are closely related to a higher-spin-symmetric Hilbert-space norm for the single-particle (super)multiplet in de Sitter space. We construct two alternative viewpoints on the type-A and type-B theories. In one view, the two co-exist in "parallel universes", specifically in even vs. odd antipodal identifications of de Sitter space. In the other view, type-A and type-B are Legendre-transformed descriptions of the same twistor-space theory in Euclidean AdS (this involves the observation that the higher-spin partition function is Legendre-dual to itself). In this way, higher-spin holography reconciles the two-boundary nature of de Sitter space with the one-boundary nature of Euclidean AdS. The supersymmetric theory that unifies type-A and type-B with bulk fermions fails to be consistent under the appropriate reality conditions.
Read the full abstract

Instrumentation and Methods for Astrophysics

Self-calibration of adaptive optics systems. Application to the THEMIS solar telescope

The interaction matrix models the effects of deformable mirror (DM) commands as perceived by the wavefront sensor (WFS) and is the cornerstone of the Adaptive Optics (AO) control. Difficulties to calibrate this matrix arise due to the number of degrees of freedom, the non-linearity, and temporal evolution of the AO system. An affine approximation of the unknown mapping between the DM commands and the WFS measurements is considered. Optimal estimators of the parameters of this affine model are obtained by fitting, in the weighted least squares sense, WFS data acquired with random probe commands sent to the DM. Several calibration methods are being considered depending on whether the model is directly fit to the WFS data or to the differences between successive WFS data. We derive closed-form expressions of the estimators of the components of the model. The proposed calibration methods can be applied under different conditions: before observing, on an internal source, or on-sky in open- or closed-loop. By introducing forgetting factors to reduce the weight of data as they age, we show that the model can be learned continuously using simple recurrence rules. The low computational complexity of these rules makes them suitable for real-time at the same frequency as the AO loop. We derive simple expressions for the mean squared errors (MSE) of the proposed estimators. We apply the proposed calibration methods to real telemetry data from the AO system of the THEMIS solar telescope. Our results show that the simple differences method is the method of choice: not only does it produce estimators with the least MSE, but it is also very simple to implement compared to the push-pull method which is widely used in AO systems.
Read the full abstract

All authors Clementine Bechet, Eric Thiebaut, Michel Tallon, Isabelle Tallon-Bosc, Bernard Gelly, Richard Douet

Programming Languages

Separating Parsing Expression Grammars using Cell-Probe Lower Bounds

We resolve three open problems concerning parsing expression grammars (PEGs). We construct a single language $C$ satisfying $C\in\mathsf{LIN}\cap\mathsf{PEG}$ and $C^R\in\mathsf{LIN}\setminus\mathsf{PEG}$. This proves that some linear context-free language is not a PEG language and that PEG languages are not closed under reversal, confirming a conjecture of Loff, Moreira, and Reis. Factoring the same witness resolves the concatenation-closure problem of Rubtsov and Chudinov negatively, in the strong form $\mathsf{PEG}\cdot\mathsf{REG}\not\subseteq\mathsf{PEG}$ despite $\mathsf{REG}\cdot\mathsf{PEG}\subseteq\mathsf{PEG}$. It also refutes closure under Kleene star, homomorphisms, and substitutions. Our main technique converts scaffolding automata (SCAs), which characterize reversals of PEG languages, into dynamic data structures in the cell-probe model. For any suitably local serialization of a problem with preprocessing, updates, and a final Boolean query, an SCA recognizer yields an exact deterministic cell-probe data structure whose operation costs are proportional to the corresponding encoding lengths. Cell-probe lower bounds can therefore prove SCA non-membership and, by reversal, PEG non-membership. We apply this transfer to Multiphase Inner Product using one-symbol update blocks and a query suffix of length $O(\log n)$, while keeping both the language and its reversal linear context-free. Ko's cell-probe lower bound then yields the witness above. The arguments are additionally formalized in Lean 4.
Read the full abstract

Computation and Language

How You Ask Shapes What You Get: A Theory-Seeded Measurement of Articulation in Advice-Seeking LLM Conversations

Users articulate the same advice-seeking request in different ways: some specify detailed constraints, others gesture at a vague need. Prior work treats this variation as noise to be averaged away; we instead treat it as a stable, measurable structure in the input distribution. We ask whether articulation (how people ask) forms latent dimensions separable from topic (what they ask about), and whether it is associated with how language models respond. We extract interpretable features from 16,447 advice-seeking prompts pooled from public chat corpora (WildChat, LMSYS, and ShareChat) and recover a small set of latent articulation factors that replicate across train/test splits and across corpora. Because this structure is largely separable from topic, the populations it defines cut across topics and stay invisible to topic- or task-based evaluation. The factors define a handful of recurring articulation styles, one of which stands out: a long-form but information-poor style, roughly one in six prompts in the largest corpus, where models return shorter, vaguer answers and do not ask for clarification even though under-specification is exactly the condition that warrants it. The contrast holds within every topic group and length quintile, and is not under-specification alone – a second, equally under-specified style does draw clarifying questions. Two independent human annotators reproduce this contrast. We argue that benchmarks should stratify on articulation, and we offer the extracted structure as a measurement instrument for doing so.
Read the full abstract

Computer Vision

Guardrail-Agnostic Societal Bias Evaluation in Large Vision-Language Models

We propose a societal bias evaluation method for large vision-language models (LVLMs) in the era of strong safety guardrails. Existing benchmarks rely on prompts that ask models to infer attributes of people in images (e.g., "Is this person a CEO or a secretary?"). However, we find that LVLMs with strong guardrails, such as GPT and Claude, often refuse these prompts, making evaluations unreliable. To address this, we change the prior evaluation paradigm by decoupling the task from the depicted person: instead of inferring person's attributes, we use prompts that do not ask about the person (e.g., "Write a fictional story about an imaginary person.") and attach the image as provisional user information to implicitly provide demographic cues, then compare outputs across user demographics. Instantiated across three tasks — story generation, term explanation, and exam-style QA — our method avoids refusals even in guardrailed LVLMs, enabling reliable bias measurement. Applying it to 20 recent LVLMs, both open-source and proprietary, we find that all models undesirably use user demographic information in person-irrelevant tasks; for instance, characters in stories are often portrayed as mechanic for male users and nurse for female users. Although still biased, proprietary models like GPT-5 show lower bias than open-source ones. We analyze potential factors behind this gap, discussing continuous model monitoring and improvement as a possible contributor for reducing bias.
Read the full abstract

All authors Yusuke Hirota, Michael Ross Boone, Arun George Zachariah, Jibin Rajan Varghese, Yu-Chiang Frank Wang, Boyi Li, Ryo Hachiuma

Artificial Intelligence

Not Safe for All: Auditing the Dialect Penalty in Text-to-Image Safety Pipelines

Text-to-image (T2I) safety guardrails fail to generalize equitably to non-standard dialects. Evaluating 23,080 paired prompts across five English dialects, we formalize this failure as the dialect penalty, where filters trigger based on linguistic surface features rather than semantic intent. Text-level filters fail in opposing directions: NSFW-T over-flags benign dialect prompts and LatentGuard over-flags toxic ones (bias gaps up to +28.29 pp), while the OpenAI Moderation API under-detects them. A controlled typo ablation confirms this penalty originates from flagging dialectal features, not generic out-of-distribution sensitivity. The pixel-level generator is largely dialect-agnostic; the penalty enters at text processing and cascades unevenly to post-hoc guardrails. We show this bias tracks training data imbalance and is mitigable via group-balanced retraining, with an ablation attributing the gain to balanced exposure rather than to the worst-group objective of GroupDRO (group distributionally robust optimization). Current pipelines systematically fail dialect speakers, an equity failure masked by mean accuracy benchmarks. Our official code and dataset are publicly available at https://github.com/minguinho26/dialect-penalty-t2i. Content Warning: This paper contains offensive, toxic, or disturbing text prompts and generated images.
Read the full abstract

Artificial Intelligence

Call Neighbours Yourself: Graph Walks with Destination-Conditioned On-Policy Self-Distillation

Reasoning over text-attributed graphs (TAGs) requires large language models (LLMs) to combine a node's text with evidence distributed across its neighbourhood. Existing methods fix the set of accessible neighbours before generation, forcing reasoning to operate over a static context and preventing the model from acquiring missing evidence during inference. We argue that neighbour selection should itself be part of the reasoning process. To this end, we propose Call Neighbours Yourself (CNY), a framework that enables LLMs to proactively explore graph neighbourhoods through topology-constrained graph-walk actions. Instead of reasoning over a pre-selected neighbour set, CNY exposes lightweight neighbour previews and learns when to expand candidate neighbours for additional evidence. To address the delayed-credit challenge of neighbour exploration, we introduce destination-conditioned on-policy self-distillation, which retrospectively evaluates a selected neighbour after its content is revealed and converts the resulting change in action preference into an action-level training signal. Experiments on standard TAG reasoning benchmarks under a unified raw-text setting show that CNY consistently outperforms fixed-context post-training baselines. Furthermore, the learned exploration policy transfers to unseen graphs and to a graph-level task not encountered during training. Code is available at https://github.com/superallen13/CNY.
Read the full abstract

All authors Yilun Liu, Boyu Luo, Yanran Tang, Ruihong Qiu, Zi Huang

Number Theory

Bloch-regulator Principal Parts of Cyclotomic Iwasawa Pseudomeasures

Let $p$ be an odd prime and let $K/\mathbb{Q}_p$ be a finite unramified extension. From a finite presentation by roots of unity of order prime to $p$, we construct a localized Iwasawa pseudomeasure on $\mathbb{Z}_p^\times$. Although the pseudomeasure depends on the chosen presentation, its image modulo bounded measures depends only on the associated Bloch class: it is the Frobenius-depleted Coleman regulator of that class multiplied by a universal half-shifted zeta principal part. Consequently, every nonexceptional weight component is bounded, while the exceptional component has at most a simple pole with explicitly determined residue. For $p>3$ and $\mathbb{Z}_p$-valued coefficients, vanishing of the principal part is equivalent to vanishing of the corresponding class in $K_3(K;\mathbb{Z}_p)$. Cyclotomic refinements preserve the Bloch class and act on the associated pseudomeasures by explicit Iwasawa multipliers. Normalized finite linear combinations of refinements interpolate arbitrary finite jets of the bounded weight-space data, subject only to the normalization at the exceptional point. We also establish a half-shifted complex Mellin factorization, compare the construction with the GSWZ germ family, and derive, at simple degree-one places above primes $p>3$, a finite-polylogarithm criterion for the local $K_3$-class of the knot $5_2$.
Read the full abstract

Materials Science

Solid and Quasi-Solid Electrolytes for Zinc Batteries: Balancing Water Activity, Ion Transport, and Interfaces

Zinc batteries offer a compelling route to safe and low-cost energy storage, yet their reliance on aqueous electrolytes promotes hydrogen evolution, corrosion, cathode dissolution, and non-uniform zinc deposition. Replacing the liquid with a solid or quasi-solid electrolyte can suppress these processes, but it also removes the medium that enables rapid Zn2+ transport and conformal electrode contact. This tension, the price of removing water, has been obscured by inconsistent use of the term solid state and by comparisons based largely on bulk ionic conductivity. Here we critically examine zinc electrolytes across a continuum from water-rich hydrogels to dry polymers, solvated crystals, and inorganic conductors. We distinguish water content from thermodynamic water activity and classify these materials according to phase state, mobile-solvent fraction, and dominant transport mechanism. We show that neither high conductivity nor nominally water-free composition reliably predicts cell performance: electrolyte thickness, Zn2+ transference, interfacial resistance, and evolving contact often determine the practical outcome. Controlled-solvation and hybrid electrolytes therefore provide the most credible near-term path, whereas genuinely solvent-free Zn2+ conductors remain a longer-term scientific target. Progress will require transparent reporting of solvent state and validation using thin electrolytes, realistic electrode loadings, limited zinc excess, and calendar-life testing.
Read the full abstract

Geometric Topology

Cyclotomic Newton Expansions and a Rank-Uniform Integer-Valued Newton Completion

Let $J_r^{SU(n)}(K;q)$ denote the reduced $SU(n)$ quantum invariant of a zero-framed knot $K$, colored by the $r$th symmetric power of the defining representation and normalized to be $1$ for the unknot. For every fixed $n\ge2$ we prove the Chen–Liu–Zhu cyclotomic expansion conjecture: there are unique coefficients $H_k^{(n)}(K;q)\in\mathbb{Z}[q^{\pm1}]$ such that \[ J_r^{SU(n)}(K;q)=\sum_{k=0}^{r} \left(\prod_{i=0}^{k-1}\{r-i\}\{r+n+i\}\right) H_k^{(n)}(K;q), \] where $\{m\}=q^m-q^{-m}$. The finite dual interpolation formula of Beliakova–Gorsky gives an integral one-sided factorial expansion. After identifying their reduced scalar with the Habiro–Lê convention, we restrict the completed center to one-row colors. Completed Harish–Chandra reflection then yields inversion symmetry in the variable $z$, and integral descent through $X=z+z^{-1}$ converts the one-sided expansion into the two-sided Newton basis. Cyclotomic-local interpolation and a UFD denominator-removal argument prove Laurent integrality of the Newton coefficients. We also determine a natural coefficient ring for a rank-uniform expansion. For every zero-framed knot there are unique Laurent differential coefficients $G_k(K;A,q)\in\mathbb{Z}[A^{\pm1},q^{\pm1}]$. The associated Newton coefficients are Laurent polynomials in $A$ over $\mathbb{Q}(q)$ whose values at every geometric node $A=q^n$, $n\ge2$, lie in $\mathbb{Z}[q^{\pm1}]$. They define a two-variable Newton inverse-limit element whose positive-rank specializations recover all symmetric-color HOMFLY–PT polynomials. The completion is taken in the Newton kernels rather than coefficientwise at roots of unity. After a positive rank and a color have been fixed, the series is finite and may be evaluated at a root of unity.
Read the full abstract

Representation Theory

Auslander-Reiten-Serre duality revisited

Let $(\mathcal{C},\mathbb{E},\mathfrak{s})$ be an extriangulated category with a right Auslander-Reiten-Serre (ARS for short) duality $(\tau, \eta)$ in the sense of Iyama, Nakaoka and Palu. We show $(\tau, \eta)$ induces right ARS dualities on the relative theories of $(\mathcal{C},\mathbb{E},\mathfrak{s})$. Under relative structures, we show that the functor $\tau$ is indeed an exact functor between extriangulated categories and preserves almost split exangles. Finally, we give some applications and examples on these results. For instance, we generalize a recent result by A. Hubery to a categorical framework.
Read the full abstract