Machine Learning
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Friday 14 August
Algorithmic Information Dynamics (AID) studies systems by perturbing them and measuring changes in algorithmic complexity, but its usual estimator, the Block Decomposition Method, is piecewise constant, restricting the calculus to finite differences. We use $K^{\mathrm{CDM}}_{\mathrm{s}F}$, a certified, differentiable estimator, to bring the calculus into learning dynamics: grokking, where a complexity order parameter is known but has not been made to act. As a transient loss kick, the estimator becomes a controller that accelerates grokking in Levin's description-length–versus-time sense, within a data-dependent Occam boundary whose finite-size trend, $f_c\sim\ln p/p$, is consistent with a coupon-collector interpretation. Ablations show that a complexity gate matches a train-loss gate in rescuing failing seeds with $27\%$ less intervention; among the tested signals, only map complexity marks the transition's completion; the certified prior and the per-parameter $\nabla K$ attribution are both fungible (a uniform-prior sensor makes bit-identical gate decisions, and random supports match $\nabla K$-selected ones above a sparsity threshold); and direct field perturbation shows a nucleation-like response to the Occam field (no linear regime is resolved over the probed amplitudes, so these measurements do not justify a fluctuation–dissipation surrogate), with a finite-field response growing by orders of magnitude toward the phase-transition. These measurements account for the empirically tuned staircase: bang–bang pulses, stall-fired and released on yield, whose iteration plausibly builds the response it exploits. The kick transfers to sparse parity and to a transformer; a sustained weight-space loss fails. The algorithmic estimator's distinct contribution is timing (when to fire and when to release), not attribution.
Artificial Intelligence
OpenAgentFlow: Enabling System-Wide Safety Boundaries for Heterogeneous AI Agent Fleets
AI agents powered by large language models are evolving from isolated assistants into heterogeneous systems in which multiple agents, planners, tools, and execution backends operate over shared environments. In such settings, safety becomes a system-level action-governance problem: deciding whether a pending action should be committed given policy-relevant state accumulated across a session. Existing safeguards operate at fragmented boundaries, making it difficult to enforce shared policies over composed action flows across heterogeneous execution paths. We present OpenAgentFlow, a control-plane/action-plane architecture that establishes the action-commit boundary as a shared enforcement interface. GUI, API, tool, and LLM-generated actions are normalized into a common AgentEvent stream and mediated by a shared pre-execution Policy Enforcement Point, while provenance, session state, audit evidence, and updatable policies are maintained outside individual agents. This provides a common governance layer across incompatible executors and allows new policies to take effect without modifying agents, prompts, models, or execution paths. We evaluate OpenAgentFlow through complementary system evaluations spanning controlled action-flow tests, a public external benchmark, policy updates, and real Android execution. On a 300-case controlled suite, OpenAgentFlow achieves 94.00% accuracy and a 95.35% attack-block rate. On the complete 1,220-case AgentDojo-Traj split of TS-Bench, it achieves 97.62% accuracy, 96.59% unsafe-action recall, and a 1.96% safe false-intervention rate. New control-plane rules take effect without modifying protected agents, and the same enforcement path operates across live GUI, API/tool, and LLM-planned Android execution. These results show that a shared action-commit boundary provides a practical basis for system-wide governance across heterogeneous agent execution paths.
Read the full abstract
Artificial Intelligence
BiasMix-Finance: Post-Generation KYC Guardrails for LLM Portfolio Advice
Large language models (LLMs) can generate plausible-sounding ETF portfolios while silently violating basic KYC-style constraints on risk, fees, and diversification. This is especially problematic in agentic multi-turn advisory systems, where each draft recommendation can become an action unless guarded by an auditable enforcement layer. We study a model-agnostic, asset-agnostic post-generation guardrail pipeline: (i) enforce a strict JSON allocation schema, (ii) validate allocations against numeric caps, and (iii) when violations occur, deterministically project the output to the nearest feasible portfolio via a convex quadratic program (QCQP). We introduce BiasMix-Finance (Mini), a compact stress-test benchmark for constrained decision-making under biased LLM generations, with a 16-ETF universe, three investor profiles, and eight bias prompts. Across three models and three inference modes (direct, critique, self-consistency), first-pass generations violate at least one cap in 47.6-85.7% of test cases (67.2% pooled), but the convex projection layer reduces final feasibility violations to 0% while requiring only a small correction distance (test pooled median D=||w*-w0||_2=0.066), indicating that the guardrail typically preserves the intent of the original allocation. We report violation rates and correction distances with confidence intervals, and paired model comparisons with multiple-testing correction. To support reproducibility, we release the dataset, prompts, caps, and code in our public GitHub repository.
Read the full abstract
Artificial Intelligence
post-graph-rag: A PostgreSQL-Native Bi-Temporal Graph RAG Engine with Temporal Grounding at Synthesis
Graph RAG connects facts no single passage states, but implementations pay three times: in infrastructure, keeping vector store, graph database and document store in sync; in quality, because a pipeline that never refuses extractor output stores edges asserting nothing; and over time, because a graph that only accumulates treats superseded and current facts alike. post-graph-rag is an open-source engine addressing all three. Chunks with embeddings, a canonical entity graph and community summaries live in one PostgreSQL database, with pgvector for search and edge tables for traversal. Extraction output is validated before writing: vague predicates, pronominal names and bare quantities are rejected, predicates normalise onto an optional vocabulary, and entities resolve to one vertex per canonical name. A bi-temporal layer records when a relation held and when the system believed it, superseding incompatible earlier assertions from document order. Against LightRAG on three corpora with extraction and embedding models fixed, it builds a denser and more queryable graph, and supersedes relationships where a baseline with no temporal model supersedes none. On LongMemEval, 500 questions of long-horizon chat memory, it scores 94.0 percent with gemini-3.6-flash against 71.2 for Zep's gpt-4o and 60.2 for full context, leading on all six question types. The largest single contribution is temporal grounding in the prompt: carrying each relation's validity period through to synthesis moves temporal reasoning from 0.496 to 0.881, ablated paired on one graph per instance. On ECT-QA, earnings-call transcripts restating every metric each quarter, it scores 0.807 under that benchmark's own protocol against 0.599, 0.406 and 0.405 published for TG-RAG, LightRAG and GraphRAG. Code: post-graph-rag https://github.com/crajah/post-graph-rag; post-graph https://github.com/crajah/post-graph
Read the full abstract
Computer Vision
Topology of a Smile: Persistent Homology in Dental Imaging
CBCT (Cone Beam Computed Tomography) scans provide detailed three-dimensional images, widely used in dentistry for diagnostic and treatment planning tasks. While invaluable, analyzing and documenting these scans is labor-intensive, prompting efforts to automate key steps like the classification and segmentation of anatomical structures to identify tooth types and associated pathologies. In this article, we propose an approach to automation that leverages persistent homology, a framework from topological data analysis that studies the shape of data by identifying features like connected components, holes, and voids across multiple scales. Persistent homology, together with a support vector machine, allows us to classify teeth in a CBCT scan and to perform diagnostics. Our method advances the state of the art, reaching average accuracy scores of 97.67% for tooth-labeling and 96.77% for diagnostic tasks, outperforming a CNN trained on the same data with accuracy of 70.27% and 86.67%, respectively.
Read the full abstract
Robotics
Evaluating Human and LLM-Generated Thematic Analysis in HRI for Vulnerable Populations: A Comparative and Ethical Analysis
Thematic analysis (TA) has long been regarded as an inherently human, reflexive, and interpretive process. However, the extent to which LLM-generated TA is appropriate for Human-Robot Interaction (HRI) research involving vulnerable populations remains largely unexamined and raises critical questions about validity and ethics, particularly in sensitive research contexts. This paper presents a comparative study of human- and LLM-generated TA in an HRI context with a focus on vulnerable populations. We evaluate both objective and semantic agreement between human- and LLMgenerated themes, and examine whether observed divergences reflect systematic interpretive patterns with ethical significance. Our analysis investigates whether LLM-generated TA risks marginalising or misrepresenting the experiences of vulnerable participants, with implications for researchers employing LLM-assisted TA in HRI.
Read the full abstract
Programming Languages
MicroPython and CircuitPython: Pythons Quiet Takeover of IoT and Robotics
Background: Python has become the dominant language in software and data science, yet embedded systems have remained tied to C/C++ due to performance and memory constraints. MicroPython and CircuitPython are changing this by bringing Python to microcontrollers, lowering barriers for IoT and robotics development. Aim: This article examines whether these platforms are achieving a quiet takeover of embedded systems, focusing on ecosystem growth, practical applications, performance trade-offs, educational adoption, and prospects. Methods: A mixed-methods design was used, including quantitative analysis of GitHub, Stack Overflow, and Google Trends data; curation of case studies from Hackster.io, Hackaday.io, and the Adafruit Learning System; and original benchmarks on ESP32 and Raspberry Pi Pico comparing MicroPython, CircuitPython, and Arduino C++ across GPIO, I2C, SPI, Wi-Fi, and memory usage. Results: Metrics show sustained growth, with MicroPython supporting over 200 boards and CircuitPython over 400. Benchmarks reveal 10-20 times slower I/O and four to six times higher memory use than C, but performance remains adequate for common sensor and network tasks. Case studies demonstrate successful deployment in home automation, robotics, wearables, agriculture, and professional prototyping. Education emerges as a primary adoption driver. Conclusions: Python is not replacing C/C++; rather, it is becoming the default prototyping and educational language for embedded systems. Continued hardware improvements, better tooling, and standardization will likely deepen this trend. The article offers balanced, evidence-based insights for developers, educators, and technology decision-makers.
Read the full abstract
Data Structures and Algorithms
A Tight Linear Deterministic Competitive Ratio for Fully Online KV-Cache Scheduling
Jaillet et al. introduced a fully online model for batching nonpreemptive LLM requests under a growing KV-cache memory constraint. For total end-to-end latency they proved that every deterministic algorithm has competitive ratio Omega(sqrt(n)), while the elementary sequential upper bound is n. We close this gap. Let R_det(n,M) be the optimal deterministic ratio for exactly n requests at memory M, and let R_det(n)=sup_M R_det(n,M). For every n >= 2 we prove (n-1)/12 <= R_det(n) <= n, so R_det(n)=Theta(n). The lower bound releases one memory-filling long request, observes its deterministic start time, and then releases n-1 wide one-token requests halfway through the long run. No short request can overlap the long one, whereas a hindsight schedule runs the two groups in the opposite order when useful. The hard instance uses the explicit fixed memory M=2(n-1)n. The upper bound is achieved by a uniform causal serial policy. The exact model, causality argument, both comparator branches, and quantifier order are machine-checked in Lean 4. Exact finite controls and replay commands accompany the proof.
Read the full abstract
Artificial Intelligence
Trust Is Not Enough: Influence Calibration for On-Policy Self-Distillation in Agentic RL
On-policy self-distillation (OPSD) gives language agents dense token-level supervision from a privileged self-teacher on the policy's own trajectories. Existing methods allocate this supervision mainly by teacher trust, but trust does not reveal whether emphasizing a token supports the current policy objective. We call this the trust-utility mismatch and introduce Influence Calibration for Self-Distillation (ICSD). For each supervised token, ICSD measures the first-order response of its importance-weighted RL surrogate contribution to a teacher-directed output perturbation. Batch-adaptive calibration converts this non-stationary signal into a bounded allocation weight while preserving the original auxiliary-loss mass within each action turn. These detached weights affect only the distillation loss and require no additional model pass. Across ALFWorld, WebShop, and Search-QA, ICSD improves all matched aggregate metrics over trust-only allocation under Group Relative Policy Optimization (GRPO) and Group-in-Group Policy Optimization (GiGPO), across two model families spanning 1.5B to 7B. At 7B, it reaches 96.1% ALFWorld success and a WebShop score of 93.1. Frozen-batch analyses show that ICSD reduces teacher-supported mass assigned to objective-opposed tokens from 60.1% to 37.8% and raises cosine compatibility with the RL gradient by 0.192. A companion repository is avail- able at https://github.com/lanqz7766/Influence-Calibration-for-On-Policy-Self-Distillation-in-Agentic-RL.
Read the full abstract
Robotics
SkillComposer: Learning Reusable Skills for Natural-Language Robot Programming
Natural-language interfaces can lower the barrier to programming robots, but existing systems struggle when users request complex tasks. While large language models (LLMs) perform well with simple commands, they often struggle to generate code for multi-step tasks, decompose high-level instructions, or reuse prior solutions. We present SkillComposer, an interactive natural-language robot programming system for simulation environments that continually learns reusable program abstractions. SkillComposer uses a generate-test architecture in which an LLM iteratively generates and revises robot programs before execution. Successful programs are stored and processed by an online library-learning algorithm that compresses recurring function sequences into reusable macro skills for future tasks. We evaluate SkillComposer through ablation experiments and a user study with 12 participants to determine its effectiveness on manipulation and robot caregiving tasks. The results show that evaluator-guided generation and learned abstractions improve success rates and usability while reducing user effort in natural-language robot programming.
Read the full abstract
Artificial Intelligence
Skill Blocks: How Should an Agent Load Its Skill? A Caching-Correct Comparison of Pre-load, On-Demand Tool-Loading, Progressive Disclosure, and Hybrid
Agent skills are often injected in full on every request, increasing token cost. We compare four content-preserving loading methods: Full, Skill Block, Reference, and Hybrid. Across SearchQA, SpreadsheetBench, ALFWorld, ScienceWorld, and SynthProc, we measure token usage using raw input for single-turn tasks and cache-correct effective input for multi-turn tasks. Results show no universal winner. Hybrid reduces input by 27.4% on SearchQA and 39.8% on SpreadsheetBench. On large multi-turn skills, Skill Block and Hybrid achieve substantial reductions, reaching 62.5% and 52.8% on ScienceWorld and 73.0% and 66.6% on SynthProc. ALFWorld shows smaller gains because procedures are short and repeatedly needed. Paired outcome tests detect no quality differences, though they do not establish equivalence. Overall, conditional loading is most beneficial when large portions of a skill are not needed on every turn.
Read the full abstract
Computer Vision
Looks Can be Deceiving: Annotator and Reviewer Performance Across Imagery Sources in Crowd-Sourced Aerial Damage Assessment
This paper presents the first known empirical investigation of annotator and reviewer performance across multi-source remotely sensed imagery, evaluating human labeling across drone, crewed aviation, and satellite views. Because existing aerial imagery datasets rely predominantly on single-source imagery, there is no currently established state of practice for efficiently allocating human labor to curate large-scale, multi-source aerial datasets. This work addresses this limitation by analyzing annotator and reviewer performance within a post-disaster building damage assessment dataset of 9 disasters, where 20041 buildings in drone, 20695 buildings in crewed aviation, and 33392 buildings in satellite imagery were labeled. These labels, provided by 187 annotators, were then refined through two successive quality-control stages: a single-reviewer pass followed by a consensus-committee review. Our analysis reveals two findings that raise questions for standard crowd-sourcing practices. First, initial annotations were revised by the final committee at rates that rise steeply from higher- to lower-resolution sources (25.27% for crewed aviation and 36.95% for satellite), with the same ordering at every observed workflow stage. Second, a single individual review reduced but did not resolve this disagreement: after review, the committee still revised 6.85% of drone, 14.05% of crewed, and 20.86% of satellite labels. These observations suggest that, in workflows like this one, uniform review allocation leaves the most residual disagreement in lower-resolution imagery. Based on this evidence, and consistent with prior work on adaptive task assignment and budget-aware quality control, this paper offers three recommendations for multi-source dataset curation.
Read the full abstract
Machine Learning
Degeneracy Counting Quantum Algorithm using Decoherence
Counting the global optima of a classical optimization problem is a #P-hard task. We develop the canonical thermal pure quantum (CTPQ) state-based degeneracy counting (CTPQsd#) algorithm that determines the number of global optima of a classical optimization problem P by measuring only a small probe S, without finding individual minima. The method exploits a perturbative relation between the decoherence measure of S and the degeneracy of P when S and P are together in a CTPQ state. We provide the first numerical demonstration that this relation can be used to count the global minima, applying it to problems encoded by diagonal random-energy Hamiltonians as a maximally unstructured testbed for classical binary optimization problems. Classical simulations of up to 20 problem qubits quantify the algorithm's sensitivity to variations in the temperature of the CTPQ state, the Hamiltonian energy range, the problem size, and degeneracy. We establish the temperature threshold for determining the exact degeneracy and identify a second, lower threshold that provides a temperature window to count near-degenerate minima within a user-defined energy tolerance. By confining measurement to S, the protocol replaces tomography over the exponentially large problem Hilbert space with tomography over a small probe represented by only four qubits.
Read the full abstract
Artificial Intelligence
When Is an Agent Evaluation Over? Outcome Finality and Cross-Unit Separation
Agent evaluations commonly score the state observed when a run stops and count the run as one trial. Interpreting that score as a final result from a separate trial requires outcome finality and cross-unit separation. Outcome finality requires that later events cannot change the claimed result, while cross-unit separation requires that earlier runs cannot change the relevant conditions of later ones. The endpoint establishes neither condition by itself, and the two can hold independently. Waiting for a delayed outcome may settle the label even though its state remains available to another run. Isolation may prevent carryover even though the scored outcome remains unresolved. We develop a completion argument that identifies the evidence needed for each decision. A final success or failure label is justified only when every relevant effect is resolved or bounded tightly enough to fix the outcome. Any remaining uncertainty must be reported. First, in a controlled replay with fixed agent actions, we find that endpoint and terminal labels differ for every nonzero-delay operation and that a delayed write changes the next run's score under shared state but has no such effect after namespacing or verified reset. Second, in a review of ten public protocols, we find that reset or deliberate retention is documented explicitly more often than unfinished operations or evidence for separate scoring. Finally, we propose an open-effects record for operations and resources that may remain relevant after the endpoint, their status, and their possible effects on the scored outcome or another run.
Read the full abstract
Robotics
From Continuous Design to Delay-Aware Discrete Synthesis: Guaranteed High-Bandwidth Joint Control for PMSM Drives
The increasing dynamic demands of modern robotic joints require current controllers to achieve high bandwidth over wide operating ranges of speed, acceleration, and torque, where communication, computation, and discrete-time effects can no longer be neglected. Conventional PMSM current controllers are typically designed in continuous time and subsequently discretized, leaving the sampling frequency and the impact of implementation delays largely to heuristic selection and iterative validation. This paper introduces a task-aware, delay-extended discrete-time joint model that explicitly accounts for physical communication and computation delays and enables direct synthesis of a discrete PI current controller with prescribed bandwidth and delay guarantees throughout the operating envelope. The framework analytically determines the minimum required sampling frequency, controller gains, and DC-link voltage needed to satisfy the specified motor and joint performance. Simulations across a range of dynamic requirements validate the methodology and demonstrate substantially reduced sampling-frequency and DC-link-voltage requirements compared with conventional continuous-time-based design. Experiments on a newly developed custom robotic joint further validate the proposed framework under real embedded implementation conditions.
Read the full abstract
Artificial Intelligence
Small Models Scout Bottleneck Order for Large-Model Data Control
Small proxy models are commonly used to identify data mixtures for larger-scale training. We ask whether their training trajectories reveal another transferable structure: the order in which larger models should resolve skill bottlenecks. We formulate first-passage skill training, where each monitored skill has a target floor and the objective is to minimize the tokens required to reach all floors. We introduce LogFloor, a closed-loop controller that directs each round toward current bottlenecks, producing phase-ordered resolution trajectories. Across five bAbI skill slices on Qwen2.5-1.5B, LogFloor reduces token cost by 56.2% on average. In 70M-to-12B transfer, three-round replay of a 70M scout path reaches every floor in all eight target runs, saving 30.9% by pair mean, 39.4% in pooled training tokens, and 37.6% under source-cost accounting. On MMLU-control, a frozen scout path succeeds across all eight 12B runs. Collapsing a path to its static marginal mixture or reversing its phase order removes most benefits, while bottleneck labels alone remain partially useful. These results identify phase-ordered bottleneck resolution as a transferable curriculum structure for monitored skill-targeted training.
Read the full abstract
Machine Learning · Cross-listed from physics.ao-ph
Developing an Offshore Machine Learning Surface Layer Scheme
Turbulent fluxes between the surface and the atmosphere are typically parameterized using empirically fit relationships. Here we test machine learning techniques for fitting the relationship for the offshore environment. To do that, data from three offshore sites are used: the Martha's Vineyard Coastal Observatory (MVCO) air-sea interaction tower, the FINO1 research platform, and the CASPER-West FLIP research vessel deployed off the coast of California. Two machine learning methods were employed: Neural Networks (NN) and Random Forests (RF). Because the observational sites had towers with measurements at different levels, the vertical differences were input as gradients. Models were built for both momentum flux and heat flux. ML models trained at the individual sites were competitive with and in some cases, better than the physically-based COARE-3 model tailored to offshore fluxes. The heat flux ML models generally outperformed the physics-based parameterizations for most metrics, but the results were mixed for momentum flux, with only the site with the most training data (MVCO) producing results better than COARE-3. When the ML models from that site were applied to the other sites, results were degraded from using data from the site being tested. ML models built from data combined from the three sites generally showed improvements for the sites with less available training data. When assessing which variables were most important, the wind speed was most important for momentum flux and temperature gradient for heat flux.
Read the full abstract
Computation and Language
Training Leaves Traces: Centered Residual Signatures for Language Model Lineage Verification
Open-weight language models are fine-tuned, quantized, pruned, and merged, yet their provenance is often undocumented. We study data-free white-box lineage verification: can weights alone reveal whether two compatible model checkpoints share ancestry? Residual training produces a shared identity-aligned component in branch products, so this structure alone cannot establish ancestry. We remove it and compare checkpoint-specific structure across residual blocks, yielding a symmetric lineage score calibrated against independent checkpoints. On residual-MLP and GPT-2 benchmarks, the score separates fine-tuned, LoRA-merged, pruned, and quantized descendants from independent and distilled models (AUROC=1.0), distinguishing weight ancestry from behavioral similarity. Under function-preserving checkpoint laundering experiments, weight-space baselines lose margin or fail; our score remains unchanged and runs 76x faster than the nearest robust baseline on GPT-2. The projection-pairing signal appears across six language-model families and beyond, and a case study correctly identifies 3 related and 7 unrelated LLaMA-2 public checkpoints. Collectively, these results establish a passive, data-free provenance signal for compatible open-weight language-model checkpoints
Read the full abstract
Artificial Intelligence
LLMs Can Predict Failure Risk, But Struggle to Predict Which Collaboration Protocol Pays Off: Cost-Aware Protocol Routing Across Reasoning Tasks
Multi-agent large language model (LLM) systems can improve reasoning by spending more computation, but deployment requires deciding when extra collaboration is worth its cost. We isolate this decision by running every problem under four protocols while holding the solver fixed within each setting: direct solving (Baseline), iterative self-correction (Single), planner-executor-reviewer collaboration (PER), and multi-agent deliberation (Broadcast). The primary benchmark comprises 4,181 competition-level math problems; paired robustness checks cover four benchmarks spanning competition math, biology, and broader science with two solver families. Across fixed policies, trained routers, and frozen LLM routers, conservative policies under-escalate, whereas higher-solve frozen routers often over-escalate. A post-answer, pre-collaboration gpt-oss-120b probe ranks Baseline failures with 0.8847 AUROC (4,151 parseable cases; 95% CI [0.8732, 0.8955]). The same score remains informative for predicting whether any collaboration helps (0.7683 AUPRC), but is much weaker for identifying PER- or Broadcast-specific value (0.1674 and 0.1041 AUPRC). Separately, the pre-answer self-confidence gate reaches 78.0% solve at 45K tokens, compared with 73.8% at 71.3K for a frozen gpt-oss-120b router and 92.4% for a retrospective fixed-order oracle. Across 10 paired model-condition settings, the oracle adds 23.2-58.3 points of retrospective coverage over Baseline, but protocol profiles vary by task. In the six settings with held-out router evaluations, oracle gaps remain 18.5-28.9 points. Confidence can therefore support initial escalation, while protocol-specific cost-aware routing remains unresolved.
Read the full abstract
Systems and Control
Geometry Induced Contraction Degradation and Stabilization of Learning Enabled Observers
Learned perception models increasingly serve as measurement maps within nonlinear observers. Their state-dependent Jacobians introduce measurement geometry that measurement-value accuracy does not constrain. Even when measurements satisfy standard accuracy criteria, the corresponding one-step Euclidean error operator can be noncontractive. Improving measurement accuracy or retuning the correction strength may leave this observer-level contraction failure unresolved. We derive the exact estimation-error operator and use it to determine whether scalar-gain adjustment can restore contraction or the correction structure must change. The analysis separates three distinct cases: a conservative certificate may be inconclusive, the nominal gain may be exactly noncontractive, or the correction geometry may lie beyond scalar-gain repair. Convexity makes this recovery boundary exactly computable. When scalar recovery remains possible, uncertainty-aware envelopes define a safe projection that changes the nominal gain only when contraction and same-state improvement are certified. Uniform local certification yields exponential error decay and explicit disturbance-dependent bounds. The method uses local composite-Jacobian information or a certified approximation and requires no retraining. Scalar experiments demonstrate certified contraction recovery and improved tracking when the geometry permits. Controlled real-image experiments provide numerical evidence of scalar-unrecoverable contraction loss despite accurate learned measurements
Read the full abstract
Computer Vision
PaSTel: Anchoring Histology in Spatial Transcriptomics via Multi-Scale Hierarchical Bio-Prior Contrastive Pretraining
Spatial transcriptomics (ST) links tissue morphology with molecular programs, motivating multimodal pretraining methods that align histology images with gene expression. However, existing approaches suffer from two key limitations: spatially informative gene selection is often dominated by ubiquitous housekeeping genes, leading to weakly discriminative representations, and independent spot-patch alignment fails to capture spatial dependencies that are critical for tissue organization. To address these challenges, we introduce PaSTel, a hierarchical multimodal pretraining framework that integrates biological priors at three levels. At the spot level, TF-IDF reweighting is used to identify spatially informative genes; at the functional level, curated KEGG pathways serve as anchors for encoding global biological semantics; and at the regional level, spatial clustering aggregates neighboring spots to model meso-scale tissue structure. Across multiple downstream tasks, PaSTel consistently outperforms existing vision and vision-omics encoders, demonstrating that incorporating multiscale biological priors yields more informative and transferable representations for spatial transcriptomics.
Read the full abstract
Computer Vision
SpIn-ViT: Designing a Sparsity-Induced Vision Transformer That Is Mechanistically Interpretable
Mechanistic interpretability has recently expanded to Vision Transformers (ViTs), with Sparse Autoencoders (SAEs) increasingly used as post-hoc tools to decompose internal representations into sparse and more interpretable features. However, because post-hoc SAEs are trained on frozen representations after the ViT has already been optimized, their latent features are not directly aligned with the downstream classification objective. We introduce SpIn-ViT, a framework that jointly trains a pretrained ViT and a modified SAE end-to-end, directly aligning sparse patch-level representations with image classification. SpIn-ViT learns semantically coherent neuron activations that localize meaningful image regions while maintaining competitive predictive performance. We evaluate SpIn-ViT across nine image-classification benchmarks using classification accuracy, quantitative interpretability metrics, AI-based and Human evaluations. Compared with the previous state-of-the-art post-hoc SAE method, SpIn-ViT achieves 8.84% higher average classification accuracy, an AI-based interpretability score nearly four times as high, and a human-evaluation score more than twice as high. We further extract interpretable rule-sets using the SAE neurons to create neurosymbolic models which achieve 5.97% higher average classification accuracy while requiring a 58.8% smaller rule-set than the neurosymbolic models created from the SOTA post-hoc SAE method.
Read the full abstract
Computational Engineering
Ground-Truth-Aware Stress Testing of a Closed-Loop Smart-Building Digital Twin Under Sensor Drift and Missing Data
Digital twins are increasingly used for smart-building monitoring and control, yet many evaluations focus on state estimation or fault detection rather than whether sensing errors materially affect closed-loop outcomes. This paper introduces a ground-truth-aware simulation framework that separates the latent physical state from a corrupted sensing layer and compares practical sensor-driven policies with an oracle controller. The synthetic twin includes 20 zones simulated at 15-minute intervals over 30 days and models occupancy-driven CO2, ventilation-energy trade-offs, sensor drift, measurement noise, and missing observations. Policy comparisons use a one-step information delay and common random numbers for paired Monte Carlo evaluation. Under nominal sensing, the raw-sensor controller disagreed with the oracle on 2.59% of decision steps, while aggregate CO2, energy, and comfort outcomes remained nearly unchanged. At 8x nominal drift, decision mismatch increased to 7.00%, but outcome gaps remained small. Across a 4x4 drift-missingness grid, none of 48 sensor-driven policy-condition combinations crossed the predeclared material-divergence thresholds. A three-sample rolling median increased nominal mismatch from 2.59% to 6.31% without meaningful outcome improvement, and Ground-Truth Regret rankings varied with utility weights. The results show that sensor error, decision disagreement, and outcome degradation are related but distinct. Because the study is fully synthetic and uncalibrated, its contribution is methodological rather than a claim of real-building performance.
Read the full abstract
Sound
Distinguishing AI-Generated Music from Edited Audio as a Hard-Negative Robustness Task
AI-generated music detectors are commonly evaluated against original songs, but real-world uploads are often remixed, re-encoded, pitch-shifted, or otherwise edited. These edited versions form a difficult negative class: they are not generated by AI, yet they may introduce spectral artifacts that resemble synthetic audio fingerprints. We study this problem as a hard-negative robustness setting for AI-generated music detection, focusing on AI-generated and edited variants derived from the same anchor songs. We compile a YouTube-based dataset of AI, edited, and original variants, using the original tracks only as references, and train a binary AI versus edited detector. Audio is processed as 10-second clips and passed as raw waveforms to a pretrained PaSST spectrogram transformer. To reduce leakage, all splits are performed by anchor song. On the held-out test set, the final video-level system achieves 0.811 balanced accuracy. At clip level, AI-generated clips reach an F1-score of 0.836, while edited clips reach a lower F1-score of 0.720. The results suggest that AI-generated music retains detectable fingerprint-like spectral cues beyond ordinary editing, but the lower edited-class F1-score shows that these cues can still overlap with artifacts from edited audio. Grad-CAM visualizations are used to inspect whether high-confidence predictions rely on localized time-frequency regions.
Read the full abstract
Numerical Analysis
Random blob methods for diffusion
Linear and nonlinear diffusion equations arise in a range of phenomena of mathematical interest, including slow and fast diffusion, sandpile dynamics, height-constrained transport, the two-dimensional Navier-Stokes equation, and dynamics for sampling probability measures. In recent years, blob methods have attracted significant interest as an approach for numerically simulating these types of PDEs. To address the $O(N^2)$ computational bottleneck of blob methods, we consider stochastic discretizations of space and time. We compare the random batch method with a new approach, which we call the random multirate method. Both of these methods build on classical stochastic methods in the optimization literature, with many similarities to stochastic gradient descent and random coordinate descent. We find that, for linear and nonlinear diffusion equations, in the tradeoff between computational complexity and accuracy, the random multirate method has the best performance. On one hand, we prove that the random multirate method converges to the underlying ODE system at a rate of $O(k \Delta t)$, matching forward Euler, and show by example that this rate is theoretically sharp. On the other hand, we observe even better rates of convergence for the random multirate method when applied to ODEs arising from blob methods for diffusive PDEs. Finally, due to its ability to simulate a wide range of nonlinear diffusion equations, including height-constrained transport and sandpile dynamics, our method succeeds in capturing key features of PDEs for which few numerical approaches exist.
Read the full abstract
Game Theory
The Open-Strategy Dictator Game: Cooperation Under Mutual Transparency
We introduce the Open-Strategy Dictator Game (OSDG), a variant of the classic dictator game in which each player's strategy is a natural-language document visible to all participants. The dictator's decision, to SHARE or TAKE an endowment, may depend on the text of the recipient's strategy. A large language model adjudicates each interaction by interpreting the dictator's strategy in the context of the recipient's. We run round-robin tournaments among diverse strategies and analyze the resulting payoff matrix using softmax equilibrium frequencies, dominance analysis, and sensitivity to the relative value of cooperation. Conditionally cooperative strategies, those that share with cooperators and take from exploiters, consistently dominate, while unconditional strategies (always share or always take) are weakly dominated. The results suggest that in environments where agents can inspect each other's decision procedures, conditional cooperation is evolutionarily robust across a wide range of payoff parameters.
Read the full abstract
Computation and Language · Cross-listed from stat.ME
Optimal Watermark Localization in Mixed-Source Large Language Model Texts
Watermarking provides a principled way to authenticate text generated by large language models (LLMs). In practice, however, the final text may be mixed-source, with watermark evidence surviving at only a subset of token positions after rewriting, insertion, deletion, or paraphrasing. Although prior work has studied global detection of watermark signals, when such signals can be localized remains unclear. We formulate watermark localization as a token-level multiple-testing problem based on pivotal statistics, with a latent indicator recording whether watermark dependence survives at each position. Under an asymptotic regime indexed by exponents for signal sparsity, next-token concentration, and effective-vocabulary growth, we derive a sharp boundary for global detection and phase transitions for discovery and classification within the class of coordinatewise pivot-based localization rules. We show that discovery is strictly harder than detection and that consistent classification is impossible across the parameter regime within this class. We then develop an adaptive thresholding method that does not require knowledge of the exponents or time-varying next-token distributions, but uses a data-driven estimate of the surviving watermark fraction. The method attains the optimal discovery boundary and near-optimal discovery power relative to homogeneous pivot-based rules. Simulations support the theoretical phase transitions, while experiments on model-generated texts demonstrate practical localization performance under common edit mechanisms.
Read the full abstract
Computation and Language
How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks
AI has long assisted scientific research, but the rapid advance of LLMs and agentic scaffolds is reshaping the landscape; a single system can now carry whole-stage research from an initial hypothesis all the way to final published paper, which is a paradigm now referred to as AutoResearch. Existing evaluations reveal little about how these agents operate or where they break down. Tasks are narrowly-scoped, evaluation measures performance but not process, and failure diagnoses lack systematic coverage or artifact-level visibility. To address this gap, we introduce AutoResearchEval, featuring 100 tasks grounded in published frontier science across 7 scientific domains and the full research lifecycle, including ideation, retrieval, execution, analysis, writing, and review. Evaluating 8 harness-model combinations yields 800 autoresearch agent trajectories, with process-level annotation. We organize these insights into AutoResearch Failure Taxonomy or ARFT, a framework of 45 empirically-grounded failure patterns. To enable scalable fine-grained attribution, we leverage a human-calibrated agent-as-a-judge pipeline to inspect complete trajectories and intermediate artifacts. Failure patterns converge on a single overarching limitation, namely that current agents lack a metacognitive loop, which entails the ability to check what they produced against what they found, revise when it does not hold up, and question whether the path they took was sound. The same patterns recur across all 8 harness-model combinations, including the strongest models tested, locating the deficit at the model level rather than in any particular scaffold; whether orchestration-level interventions can close it is an open question this work does not test. We publicly release AutoResearchEval and ARFT to facilitate continued research and development in autonomous scientific discovery.
Read the full abstract
Artificial Intelligence
Frontier AI Forecasting Has a Measurement Problem: An Audit of Progress Evidence
Quantitative forecasts of frontier artificial intelligence often connect dated targets to trends in benchmark scores, training compute, release time, or expert belief. This paper audits whether the public measurement record supports those connections before another trend is fitted. I construct a frozen, event-centric record through 12 August 2026 with 62 selected systems, 12 versioned benchmarks, seven capability or impact criteria, 144 graded events, 27 source records, and 408 typed relations. The record is an audit sample, not a census. Only seven systems jointly observe estimated training compute and a METR 50 percent task horizon. Training compute is absent for 19 of 27 closed systems, including every selected closed release from 2026, while none of the 35 open-weight systems has a METR horizon observation. Benchmark succession creates a second break: a seven-system link from METR Time Horizon 1.0 to 1.1 has a log-scale slope of 1.206 (95 percent CI 1.021 to 1.390), whereas a six-system MMLU to MMLU-Pro comparison appears shift-like under logit and probit links but not under linear or logarithmic links. The observed bridges have about 80 percent power only for slope departures near 25 percent. Provenance is concentrated: 52 of 71 substantive quantitative events, or 73.2 percent, come from one measurement programme, and 76.1 percent are laboratory releases. A review of 56 methodological and empirical sources identifies 16 complementary measurement directions spanning resources, inference budgets, reliability, agentic work, safety, human preference, field outcomes, and forecast backtesting. No direction supplies a replacement scalar. The result is not that frontier AI forecasting is impossible, but that a defensible dated forecast is a claim about a versioned measurement system with explicit joins, protocols, links, and source dependence, not merely a fitted curve or calendar date.
Read the full abstract
Robotics
Geometry-Aware Online Mapping for 3D Gaussian Splatting SLAM
Recent 3D Gaussian Splatting (3DGS) has enabled efficient photorealistic view synthesis and is rapidly being adopted in simultaneous localization and mapping (SLAM) systems for online mapping. In these systems, a Gaussian map must be expanded and refined incrementally while tracking runs in real time, so initialization and density control directly determine where limited computation and iterations are spent. This contrasts with offline 3DGS reconstruction, where such heuristics can be amortized over long optimization schedules. However, most 3DGS-SLAM pipelines inherit initialization and density-control heuristics from offline reconstruction, which can become brittle under the strict per-keyframe optimization budgets and incremental map growth of online SLAM. In this work, we revisit these heuristics in a decoupled 3DGS-SLAM setting and propose three geometry-aware methods that operate in the mapping thread: transmittance-preserving densification, camera-aware scale initialization from depth and intrinsics, and error-guided densification that focuses new primitives on high-residual regions. Our results show consistent improvements in rendering quality with negligible overhead, highlighting the coupling between photometric residuals and pose uncertainty in online SLAM. We will open-source our code to the community to foster growth and validate reproducibility.