arXiv:2606.03694v2 Announce Type: replace Abstract: Meaningful human-robot interaction (HRI) requires a robot to continuously assess user engagement through persistent user tracking. However, state-of-the-art Multi-Object Tracking models are heavily optimized for surveillance or autonomous driving. A social robot faces distinct egocentric challenges, such as humans moving in unpredictable nonlinear patterns, obstructing each other, or leaving and reentering the scene. These dynamics trigger frequent identity switches (IDSW), causing the robot to lose its footing mid-conversation. To address this, we introduce a focused, custom-annotated egocentric dataset collected via the Furhat robot. We present a systematic evaluation isolating detection errors from tracking logic, comparing face versus body tracking, and assessing the impact of extended memory and appearance re-identification (ReID). Results indicate that increasing temporal memory mitigates prolonged occlusions but fails on complex dynamic events. Integrating ReID resolves complex switches but exhibits opposing effects: it substantially improves body tracking stability, yet causes facial IDSW to spike due to profile angle sensitivity. Ultimately, our optimized pipeline reduces IDSW by 49% compared to a standard tracking-by-detection baseline, effectively mitigating interaction breakdowns. As standard benchmarks lack dense, close-quarter occlusions, this work highlights the critical need for natively captured social dynamics to truly validate HRI perception models.
Science Journals
arXiv:2606.04068v2 Announce Type: replace Abstract: This study presents a unified nonlinear dynamical framework for understanding, modelling, and teaching competitive judo. The Tori Uke Dyad is formalised as a constrained multibody system whose behaviour emerges from symmetry breaking, coupling dynamics, and transitions between attractor basins. Two fundamental instability archetypes rotational collapse Uchi mata type and gravitational lever collapse Seoi otoshi, suwari version are identified as the core pathways through which all throwing techniques evolve. A Functional Instability Index It is introduced as a dimensionless order parameter. Derived from local bifurcation analysis, it integrates geometric, dynamic, and coupling related variables through a multiplicative nonlinear structure, enabling the quantification of critical transitions such as Kuzushi, Tsukuri, and Kake. Fractional Brownian Motion models the global displacement of the Dyad, where the local H\"older exponent encodes the informational structure of the interaction. An AI based pipeline extracts instability signatures from high frequency competition video, providing objective measures such as finite time Lyapunov exponents, attractor topology, and coupling stiffness. Building on these principles, a three level teaching framework is proposed, shifting judo pedagogy from a technique centred to an instability centred approach.This study establishes the first theoretical foundations for a predictive science of judo performance and outlines future directions for empirical validation, athlete monitoring, injury risk modelling,cross sport applications and judo as neurological rehabilitative tool for Parkinson diseases.
arXiv:2606.04764v2 Announce Type: replace Abstract: Whether attention maps from pathology foundation models capture genuine biology remains unknown, yet this question is critical for clinical trust and regulatory approval. We propose a spatial transcriptomics-based framework for orthogonal, hypothesis-free evaluation of attention and apply it to five pathology foundation models (CONCH v1.5, UNI v2, Virchow2, GigaPath, H-Optimus-1) and a ResNet50 baseline. Using attention-based multiple instance learning, we train single-task and multi-task models to predict five molecular alterations in glioblastoma on the CPTAC cohort, validate on an independent TCGA cohort, and evaluate biological coherence of attention maps against 87 transcriptional signatures using co-registered Visium spatial transcriptomics data from 18 samples. Internally, no single encoder dominates across all tasks, and external validation inverts internal performance rankings. Attention maps show a five-fold enrichment gradient from pathways (Cohen's d=0.329) to individual genes (d=0.055), indicating that attention captures emergent multi-gene transcriptional programs rather than individual molecular events. Spatially smooth attention maps do not imply biological coherence, and different encoders attend to distinct biological compartments. Our framework provides objective, quantitative assessment of what foundation models learn from histopathology, moving the field beyond qualitative saliency map review.
arXiv:2606.04883v3 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used in workflows for generating formal proofs in Lean. These workflows often decompose problems into smaller lemmas, sample many proof attempts, and use compiler feedback to guide search. However, they can be prohibitively expensive, often spending substantial compute on attempts that ultimately fail. In this work, we address this problem with an action routing agent that consists of a data plane and a control plane. The data plane generates natural-language lemma decompositions, formalizes them in Lean, and samples proof attempts for the resulting theorem and lemma targets. The control plane observes previous failed Lean attempts, estimates both the likelihood of success and the cost of another attempt, and decides whether to continue proving the current target or restart from a new breakdown. On a subset of PutnamBench, our agent decreases the cost by 28.9% over a fixed-step baseline on average, preserving performance while using substantially less compute. These results suggest that failed Lean trajectories provide actionable signals for cost-aware resource allocation in agentic theorem proving.
arXiv:2606.05347v2 Announce Type: replace Abstract: Every new clinical imaging device creates a domain shift where dense gland masks are expensive yet cheap clinical signals -- eyelid outlines, Pult grades, morphometric ratios -- are routinely recorded. We present TopoPult-SSL, a two-stage framework for cross-device meibomian gland segmentation. Stage 1 adapts a source-trained model without target gland masks in the training loss, using four weak-prior anchors driven by target eyelid masks and clinical metadata only. Stage 2, when target gland masks are available, distils complementary Stage-1 teachers into a single compact student via supervised self-distillation. We develop and validate the technique on the public MGD-1k to CAMG research benchmark (1,000 to 100 images, different device), where the distilled model achieves Dice 0.716+/-0.006 (best 0.726), surpassing UA-MT (0.710) and the ensemble teacher (0.720) -- with a single pass. The gland-mask-free Stage-1 variant reaches Precision 0.694 vs. 0.30-0.34 for SAM/MedSAM (p<0.001), enabling deployment without dense gland contouring. Code and reproducibility scripts are released.
arXiv:2601.02272v2 Announce Type: replace Abstract: In a staged, beam-driven, plasma wakefield accelerator, electrons are accelerated in a sequence of plasma stages, each powered by a driver electron bunch. Between each stage, a magnetic chicane is used to dispose of the spent driver and to inject the new fresh one, while transporting the witness bunch. We discuss the effect of synchrotron radiation in the interstage section on the accelerating gradient and on the final energy reach of such a machine.
arXiv:2606.05442v2 Announce Type: replace Abstract: For non-interacting fermions in a harmonic trap, the partition function at any discrete number of imaginary time slices (or beads) and for any choice of short-time propagator admits an exact recursion relation derived directly from the contracted determinant form of the path integral. This finite-bead recursion is distinct from earlier continuum-limit recursions, which do not apply to the discrete time partition functions. By identifying a direct correspondence between this recursion and Newton's identity, application of a closed-form result from the theory of partitions provides an exact expression for the one-dimensional $n$-fermion finite-bead partition function. From this, the Thermodynamic and Hamiltonian energies and specific heats are analytically calculated for any $n$, $N$, $\tau$, and propagator choice.
arXiv:2606.05896v2 Announce Type: replace Abstract: Creating lifelike digital humans with genuine social intelligence requires unifying cognitive reasoning and multimodal generation within a coherent framework. Current approaches treat these as separate tasks: Large Language Models excel at dialogue but lack embodied expression, while diffusion-based talking head models achieve visual fidelity but ignore social cognition. To bridge this gap, we propose a closed-loop dual-agent framework integrating perception, social reasoning, and expression into a continuous interaction cycle. The perception module analyzes partners' multimodal behaviors from video, while the social reasoning module infers hidden mental states through Theory of Mind and selects responses via an ensemble mechanism. The expression module then generates emotion-controllable videos that jointly synthesize speaker speech and facial expressions with listener reactive behaviors, capturing bidirectional dynamics absent in prior work. We further construct a hierarchical Persona-Scenario dataset with psychologically grounded personas and private social goals to support evaluation under information asymmetry. Experiments on this dataset demonstrate competitive or superior performance on both dialogue quality and video generation metrics. Notably, our method surpasses even the full-information Script mode on key dialogue quality dimensions, suggesting that explicit mental state inference under uncertainty can elicit more thoughtful dialogue than unrestricted information access. Project page: https://resonantminds.github.io/.
arXiv:2601.04524v2 Announce Type: replace Abstract: Understanding biomedical experiments provides a foundation for downstream tasks, e.g., laboratory automation, and facilitates effective cross-disciplinary communication. Two challenges, High Information Density (HID) and Multi-Step Reasoning (MSR), pose unique difficulties for precise experimental understanding. Extracting structured knowledge, e.g., Knowledge Graphs (KGs), is an effective approach to address the HID and MSR. However, existing biomedical datasets for structured knowledge information extraction are limited to a general or coarse-grained level, hindering fine-grained experimental understanding. To address this gap, we introduce Biomedical Protocol Information Extraction Dataset (BioPIE), a dataset providing procedure-centric KGs that capture entities, actions, and relations at a scale sufficient for reasoning across biomedical protocols. We evaluate information extraction methods on BioPIE and implement a question answering system leveraging the dataset for validation, demonstrating improved understanding performance on test sets as well as on the HID and MSR question sets.
arXiv:2606.07555v3 Announce Type: replace Abstract: Glossaries, technical specifications, and system prompts routinely ask language models to use familiar words in unfamiliar ways. When this works, the local rule does not overwrite the old meaning; the pretrained prior keeps operating underneath, and its strength still shows through. We test this with a Stroop-style paradigm: a remapping rule (doctor means forest) pitted against the query word's lexical-prior distractor (hospital), with matched neutral controls. Across 11 open-weight models spanning four families and 1B-9B parameters, lexical-prior strength predicts interference even after item-level controls for answer prior, frequency, tokenization, and prompt wording. Activation patching on five models then locates where the override is repaired internally. Restoring three source positions that carry the redefinition (the definition subject, its new target, and the query word) almost fully recovers the effect (aggregate $R \in [0.92, 1.06]$). The repair works by protecting the contextual target rather than by silencing the prior; the distractor's probability falls whenever these positions are perturbed, but the target survives only when the redefinition is restored intact. Behavior and mechanism converge on the same channel: the prior's strength both predicts which overrides fail and marks where the causal repair lands.
arXiv:2602.11798v2 Announce Type: replace Abstract: Next-generation communication networks are characterized by integrated ultra-high reliability, ultra-low latency, massive connectivity, and ubiquitous coverage. However, this paradigm faces significant structural challenges of liquidity and security. Liquidity issues arise from prohibitive upfront costs of network resources, which strain the limited capital and financial flexibility. This also limits the deployment of the resource- and investment-intensive security solutions, bringing security issues. Security vulnerabilities arise from the decentralized architecture as well, particularly threats posed by Byzantine nodes. To address these dual challenges, we propose a novel framework utilizing Real-World Asset (RWA) tokenization for tokenizing network resources. RWA tokenization uses blockchain to convert ownership rights of real-world assets into digital tokens that can be programmed, divided, and traded. We then analyze the criteria for identifying suitable assets. Through a case study on dynamic spectrum allocation, we demonstrate the superior performance of this RWA approach. Particularly under conditions of resource scarcity, it can exhibit strong resilience against collusion and default attacks. Finally, we delineate fruitful avenues for future research in this nascent field.
arXiv:2606.23768v1 Announce Type: new Abstract: We propose cryptographic certificates of validity for agentic AI systems. The core idea is to formally specify a correctness or policy condition as a logical predicate, compile this predicate to a witness-checking problem over polynomial constraints, and use a succinct cryptographic proof system (and optionally zero-knowledge) to certify that the condition holds. This offers a middle ground between formal verification of source code, and cryptographic authentication. An agent's action can be accompanied by an independently checkable proof that it satisfies an agreed formal policy, without requiring the verifier to trust the agent or to re-execute computation. We outline the approach at a high level, give the core mathematical translation, relate the proposal to proof-carrying code, zkVMs, formal methods, and agent governance, and note the specification, auditing, and deployment questions that a full implementation must answer.
arXiv:2606.23770v1 Announce Type: new Abstract: Entropic Dynamics (ED) provides a statistical-inferential foundation for physical laws, deriving motion and field equations from principles of entropy maximization rather than quantization postulates. ED reconstructs quantum mechanics by treating the evolution of probability distributions on configuration space as driven by information constraints, yielding the Schrodinger equation as a non-dissipative diffusion process. Building on this foundation, the present work extends the ED framework into a Unified Entropic Dynamics (UED) formulation that encompasses classical, quantum, relativistic, thermodynamic, and gravitational phenomena within a single information geometric principle. By maximizing entropy subject to constraints on diffusion, drift, and gauge covariance over a manifold endowed with a supermetric H_ab, we derive a universal field equation that merges the Fokker-Planck and Hamilton-Jacobi structures into one covariant form. When specialized to different dynamical variables, this equation reproduces the harmonic oscillator, Schrodinger, Maxwell, Klein-Gordon, and gravitational wave equations, thereby revealing a deep equivalence between probabilistic inference and dynamical law. The UED framework demonstrates that spacetime geometry, quantum coherence and thermodynamic diffusion emerge as complementary expressions of the same entropic process, establishing a unified inferential foundation for both microscopic and macroscopic physics. In this formulation, energy, probability, and entropy are intertwined aspects of information geometry, providing a consistent inferential foundation for understanding classical, quantum, and gravitational dynamics as complementary expressions of a single entropic law.
arXiv:2606.24391v1 Announce Type: new Abstract: We introduce Age of LLM, a turn-based 1v1 benchmark in which two LLMs face off on a 13x7 grid to destroy the enemy base. Three stressors are deliberate: fog of war, full diplomacy (messages, ceasefires, ultimatums; uranium kept secret), and a reliability dimension where every turn must follow a strict JSON schema and an illegal action is silently discarded. The engine is private and each match uses a fresh random map seed and opponent, mitigating the data contamination that affects public benchmarks. Models receive a (near) rule-only prompt with no build-order advice (two tactical seed phrases were present during data collection; see Section 2.7). We benchmark 15 reasoning models across 54 matches and 5,258 actions. Findings: (1) the nuclear rush dominates (78% on the rules-coherent v0.11+ sub-corpus; 85% corpus-wide) with a sole-launcher signature that is largely mechanical under secret-simultaneous launch rules, not a cognitive deterrence failure; (2) military conquest is rare but faster (12.3 vs 18.9 turns); (3) diplomacy is prolific yet almost never consummated; (4) ~58% of illegal actions are fog/state errors, making the illegal-action rate a measure of belief-tracking; (5) -- the least established, and the only one we label exploratory -- a weak link associates reliability with winning. The corpus is small, unbalanced and not side-swapped, so the ranking is a preliminary descriptive view, not a contribution. Beyond ranking, the turn-by-turn traces of actions and messages make the corpus a lens on how LLMs reason under adversarial uncertainty -- their belief-tracking, spontaneous deception, and per-model cognitive "personas" -- which we frame as a future research direction. We release the replay format, an isometric viewer and all replays; engine source on request.
arXiv:2606.24453v1 Announce Type: new Abstract: Modern coding agents pair LLM generators with various tools, including cheap diagnostics and expensive verifiers. The tool-use decisions are typically governed by orchestrators that often use fixed rules and ignore uncertainty. We formulate orchestration as cost-sensitive sequential hypothesis testing: a Bayesian controller maintains a belief over candidate correctness and dynamically decides whether to gather more evidence, refine the candidate, verify it, or stop. Across six generators and nine coding benchmarks, Bayesian control proves to be most valuable when verification is costly and critics are informative but imperfect. Beyond control, the belief state yields an interpretable correctness score that outperforms token-probability and raw tool-success baselines for uncertainty quantification.
arXiv:2606.24129v1 Announce Type: new Abstract: For a wheelchair user, a standard blue line on a map is often a broken promise. While platforms like OpenStreetMap (OSM) successfully capture where a path is, they frequently fail to convey how it physically feels to travel on it. This information barrier is problematic for wheelchair users. To solve this issue, we present OmniPath, a system that moves from passive mapping to proactive environmental auditing. Our framework fuses the network topology of OSM with the submeter precision of high-density aerial LiDAR (USGS 3DEP) to create a high-fidelity 3D model of the pedestrian environment. Rather than simply routing a user, our agent virtually traverses the network, analyzing the surface in 0.5 meter increments. It rigorously quantifies physical friction points specifically running slope, cross slope, and vertical discontinuities against ADA compliance standards, calculating a weighted severity score to categorize hazards from ``Mild'' to ``Critical.'' To ensure real world reliability, we validated the system against 200 physical ground truth field surveys across the National Mall using stratified random sampling. The framework demonstrated strong diagnostic reliability for high-severity hazards, achieving F1-scores of 0.60 for Severe and 0.58 for critical categories. By automating this micro-scale inspection, OmniPath identifies the ``invisible'' barriers that standard maps miss, effectively transforming a static dataset into accessibility data source that anticipates accessibility challenges before the user ever leaves home.
arXiv:2606.24151v1 Announce Type: new Abstract: Self-evolving agents improve over time by distilling experience from past executions and reusing it in future tasks. Existing systems represent such experience either as natural-language text injected into the agent context or as code exposed as callable tools. However, the choice between these representations is typically made at design time rather than derived from the characteristics of the experience itself, leaving the trade-offs between them poorly understood. We present the first controlled study that isolates text memory and code memory over an identical set of experiences. Our results show that the two forms exhibit complementary trade-offs in construction cost, execution efficiency, and transferability, such that neither representation alone is sufficient. Guided by these findings, we propose Metis, a self-evolving agent system built on a hierarchical dual-representation memory. Metis organizes textual experience into execution plans, environment facts, and common pitfalls, and selectively crystallizes recurring plans into validated callable tools. This design combines the broad applicability of text memory with the execution efficiency of code memory while incurring tool-generation cost only when justified by repeated reuse. We evaluate Metis on AppWorld, a challenging benchmark for interactive agents. The results show that Metis improves task accuracy by up to 20.6% over ReAct while reducing execution cost by up to 22.8%. Compared with representative self-evolving agent systems, Metis consistently achieves a better balance between accuracy, execution efficiency, and memory-construction cost.
arXiv:2606.24367v1 Announce Type: new Abstract: Auditory models are central tools for studying spatial hearing, yet their validation typically relies on heuristic performance metrics rather than principled statistical methods. We present two contributions building on a Bayesian sound localisation model that jointly infers sound direction from noisy perceptual features and individual head-related transfer functions (HRTFs). First, we derive an explicit likelihood function and validate it through parameter recovery on simulated data and fitting to behavioural responses from 33 participants, demonstrating that the framework reliably identifies individual sensorimotor and spectral parameters. Second, we use this framework to compare four HRTF template interpolation methods, showing that full-sphere spatial coverage and high-frequency spectral fidelity are the primary determinants of template quality, while the specific interpolation algorithm is secondary. Together, these results show that standard model-based statistical methods can address both fundamental questions in spatial hearing and applied problems such as perceptual HRTF evaluation. An open-source Python implementation is released alongside this work.
arXiv:2606.24377v1 Announce Type: new Abstract: Compliant joints can embed safety and adaptability into dexterous hands, but achieving large-stroke anthropomorphic motion while maintaining joint-specific, directiondependent stiffness and reliable proprioception remains challenging. This paper presents the PDS joint, a parametric doublespiral (PDS) compliant joint that enables systematic shaping of directional stiffness across multiple deformation modes, including flexion/extension, abduction/adduction, and pronation/supination. We instantiate the joint using Archimedean and logarithmic spiral templates for different hand joints and introduce an asymmetry ratio to tailor stiffness distributions for both grasp stability and hyperextension resistance. To make the joint practically usable under large deformation, we co-design embedded inductive proprioception and propose a learningbased calibration pipeline that maps raw inductive signals to joint states using ArUco-marker tracking. Experiments characterize the stiffness landscapes across geometric parameters and demonstrate a non-monotonic dependence of lateral support on asymmetry, indicating the importance of principled parameter tuning. For joint-state estimation in the most challenging abduction/adduction motion, a learned multilayer-perceptron (MLP) mapping reduces the error compared with conventional curve fitting by 41.6%. Finally, we integrate the proposed joints into an open-source dexterous hand as a demonstration platform, on which the hand grasps a set of nine everyday objects and performs safe, contact-rich human-involved interactions.
arXiv:2606.07560v2 Announce Type: replace Abstract: Function-vector (FV) heads are identified by the magnitude of their causal contribution to in-context rule tasks, and the resulting top set is treated as a single functional class. We show this hides a sign structure. Under a sign-preserving criterion (refined direct logit attribution, validated head by head with path patching) the FV population splits into two opposing groups: writers push the rule-correct logit up, cancellers push it down, and ablating both together moves the readout less than the sum of the two. The split is causal and reproducible. It holds in all but two of the fifteen (model, task) cells we test, spanning three architectures and six Pythia scales, and a sign-shuffle null rejects the single-class account in all but one of the six main cells. It is also invisible to magnitude-only ranking, which surfaces whichever group locally dominates and misses the other, so any function vector or ablation built that way silently averages a promoting and a suppressing mechanism. Cancellers are not attention sinks, induction heads, or copy-suppression heads, and their causal effect is larger than that of magnitude-matched non-FV controls. Zero-ablating them recovers $+0.13$ to $+0.29$ nats on the correct label in every main cell, and shifts accuracy by $+2$ to $+7$ pp in the same direction.
arXiv:2606.09313v2 Announce Type: replace Abstract: Retrieval algorithms are used to estimate atmospheric concentrations of greenhouse gases (GHGs), such as carbon dioxide (CO2) and methane (CH4), by solving inverse problems from high-spectral-resolution satellite radiance measurements. However, these algorithms are computationally expensive, which makes real-time estimation at scale difficult. Machine-learning models have therefore been proposed as fast emulators of retrieval algorithms. Most existing studies, however, evaluate them only on test data from the same period as the training data. We study the stability over time of such emulators using data from the Greenhouse Gases Observing SATellite (GOSAT). We show that prediction accuracy generally deteriorates when the test period moves away from the training period. We also show that including time as an input feature substantially improves XCH4 prediction for Lasso and neural-network models. Among the methods considered, a simple Lasso model performs as well as or better than more complex methods such as neural networks, and yields more stable predictions over time. We further validate the results using the Total Carbon Column Observing Network (TCCON), a ground-based observation network. On the TCCON-matched dataset, the time-augmented Lasso achieves errors against TCCON that are comparable to the disagreement between GOSAT and TCCON for both XCO2 and XCH4.
arXiv:2606.09824v4 Announce Type: replace Abstract: Similarity search is a fundamental operation in time series analysis. Most existing techniques, however, require users to supply a precise sequence of values (typically an entire time series object) as the query input. This rigid requirement limits real-world applications, where users instead want to express patterns, trends, or value ranges. Flexible, pattern-based search has been explored in text retrieval and complex event processing, but remains underexplored for large-scale distributed time series. To close this gap, we propose TSseek, a regular-expression-powered search framework for distributed time series datasets. TSseek's query language enables users to compose patterns encompassing trends, value ranges, and wildcard segments. We show that conventional approximation techniques (e.g., PAA and SAX) and their index structures are ill-suited for such queries because they cannot operate on regular-expression query constructs. In TSseek, we map the time series objects and the query constructs into the same space by approximating time series objects as sequences of line segments that retain both trend (slope direction) and value range, and translating query constructs into bounding rectangles. To support efficient processing, we build TSseek-X, a distributed spatial index over the time series segments. TSseek supports two fundamental query types, namely whole-matching queries (over entire series) and subsequence-matching queries (over arbitrary windows within a series). Across benchmark and real-world datasets, full-scan, model-based, and SAX-based baselines all sacrifice either accuracy or speed, whereas TSseek returns exact answers efficiently. Also, for subsequence workloads, it achieves significant speedups over state-of-the-art subsequence matching engines.
arXiv:2606.24423v1 Announce Type: new Abstract: We develop a sparse spectral method for solving partial differential equations on a class of two-dimensional geometries bounded by algebraic curves. The numerical method uses generalised bivariate Koornwinder polynomials which form a complete orthogonal basis, but one which is not graded in terms of polynomial degree. The polynomials are built from new families of univariate semiclassical orthogonal polynomials whose associated operator matrices (Jacobi matrices, raising matrices and differentiation matrices) are computed with optimal linear complexity in the number of basis functions. When used to discretise partial differential equations the resulting matrices are sparse enabling efficient numerical solution. Moreover, we develop fast transforms from values on a grid to expansion coefficients. The efficiency and accuracy of the resulting spectral method are illustrated through a series of numerical experiments on geometries whose boundaries are smooth and piecewise smooth including non-convex geometries. We observe algebraic convergence for geometries with corners, which accelerates to exponentially fast (spectral) convergence when the boundary is smooth.
arXiv:2606.24381v1 Announce Type: new Abstract: Prompt-based interaction has become a dominant paradigm for using large language models (LLMs), where multiple candidate prompts are evaluated and the top-ranked one is selected for downstream use. This workflow implicitly assumes that prompt rankings are stable under minor variations in evaluation conditions. In this paper, we systematically study prompt ranking stability under common sources of variability, including random seeds and limited evaluation subsets. Across three open-weight LLMs and two benchmark tasks, we find that while overall rank correlations are often moderate to high, the identity of the top-performing prompt frequently changes, leading to unreliable selection decisions. To address this issue, we propose a simple stability-aware selection strategy based on a lower confidence bound, which accounts for both performance and variance. Our results show that this approach improves robustness in unstable settings while remaining competitive in more stable regimes. These findings highlight the importance of accounting for evaluation uncertainty in prompt selection and LLM benchmarking.
arXiv:2606.23890v1 Announce Type: new Abstract: To date only one chiral species, propylene oxide, has been observed in the interstellar medium but little is known about the chemistry that leads to a detectable abundance of this molecule in Sagittarius B2. We used a glow-discharge ion source and a room-temperature ion trap to study the neutralization reactions necessary to convert propylene oxide cation (PO$^+$) -- the assumed precursor for propylene oxide in space -- into the observed astrochemical. We found that the charge-transfer reaction between PO$^+$ and ammonia (NH$_3$) proceeds with a pressure-independent rate coefficient of $(1.39\pm0.03)\times10^{-12}$ cm$^{3}$ s$^{-1}$ to neutralize PO$^+$ and form the radical cation NH$_3$. Although this measured rate coefficient is much slower than that predicted by capture theories, the high abundance of NH$_3$ in Sagittarius B2 motivates the inclusion of this reaction in astrochemical models. We hypothesize that the low ionization energies of many chiral molecules important to origin-of-life theories means these species may exist as cations in the interstellar medium.