Papers

Recursive Super-Resolution (SR) extends fixed-scale SR to extreme magnification by repeatedly feeding predictions back into the same model, analogous to zooming an image repeatedly. However, ground truth availability at every scale, especially at depth, remains challenging as the required source resolution grows geometrically, leaving deeper predictions unsupervised. We present OracleZoom, an on-policy distillation-inspired, reference-constrained framework that trains on its trajectory while carrying the last ground-truth evidence beyond the supervision boundary. Direct and cross-scale supervision constrain verifiable content, while a no-reference quality objective guides unresolved fine-scale detail. A KL-constrained pretrained latent prior limits quality-driven drift, while EMA consistency stabilizes the supervision boundary. Across seven datasets, OracleZoom achieves the state-of-the-art SR quality across zooming scales, averaging 0.713 CLIPIQA, with larger gains on deeper scales, while significantly reducing hallucinations. Code, data, and models are available at https://dipta007.github.io/OracleZoom/ .

Monte Carlo Tree Search (MCTS) has demonstrated success in online planning for deterministic environments, yet significant challenges remain in adapting it to stochastic Markov Decision Processes (MDPs), particularly in continuous state-action spaces. Existing methods, such as HOOT, which combines MCTS with the Hierarchical Optimistic Optimization (HOO) bandit strategy, address continuous spaces but rely on a logarithmic exploration bonus that lacks theoretical guarantees in non-stationary, stochastic settings. Recent advancements, such as POLY-HOOT, introduced a polynomial bonus term to achieve convergence in deterministic MDPs, though a similar theory for stochastic MDPs remains undeveloped. In this paper, we propose a novel MCTS algorithm, \Algname, designed for continuous, stochastic MDPs. \Algname integrates a power mean as a value backup operator, alongside a polynomial exploration bonus to address the non-stationarity inherent in continuous action spaces. Our theoretical analysis establishes that \Algname converges at a polynomial rate of $\mathcal{O}(n^{-\zeta})$, $\zeta \in (0,1/2)$, where \( n \) is the number of visited trajectories, thereby extending the non-asymptotic convergence guarantees of POLY-HOOT to stochastic environments. Experimental results on stochastic tasks validate our theoretical findings, demonstrating the effectiveness of \Algname in continuous, stochastic domains.

Planning with a generative model aims to estimate the value of a state using as few simulator calls as possible. SmoothCruiser achieves problem-independent complexity $\widetilde O(\varepsilon^{-4})$ by exploiting the smoothness of the entropy-regularized Bellman backup, but its estimator is only first-order. We show that the sample-complexity exponent of SmoothCruiser-type planners is governed by the order $\beta$ of the local Taylor remainder, giving oracle complexity $\widetilde O(\varepsilon^{-(2+2/(\beta-1))})$: the first-order case $\beta=2$ recovers SmoothCruiser, while a second-order/cubic remainder $\beta=3$ yields $\widetilde O(\varepsilon^{-3})$. We reach this regime with an optimal-transport-smoothed Bellman backup over action distributions, which has a closed form, a policy gradient, and a Lipschitz Hessian, and whose quadratic correction admits an unbiased cross-product estimator. The resulting SecondOrderSmoothCruiser achieves $\widetilde O(\varepsilon^{-3})$ oracle complexity for fixed OT parameters, and we relate the OT, entropy-regularized, and unregularized objectives through explicit regularization-bias bounds.

Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to navigate unseen environments by following natural language instructions. Current zero-shot VLN-CE methods either rely on pre-trained waypoint predictors or require multiple queries to large models per step. To address prohibitive inference latency and computational overhead, we propose O2C-Nav, an efficient zero-shot navigation framework that calls only a single large model once per decision step. Our approach introduces a training-free structured waypoint generator and a novel abstract representation that projects sparse, history-aware candidate waypoints directly onto RGB images as visual markers. The MLLM selects a waypoint or generates a fallback target bounding box at each step, while a low-level Fast Marching Method (FMM) planner converts the selected target into an executable collision-free path. This paradigm provides the model with concrete spatial perception and explicit memory while significantly reducing the visual processing load. Extensive evaluations on the R2R-CE and RxR-CE benchmarks demonstrate that O2C-Nav outperforms current state-of-the-art zero-shot methods, highlighting its great potential for real-time robotic deployment. Code is available at https://github.com/kkpsq/O2C-Nav-Code.

A provider of a high-risk AI system must keep records that make a decision traceable, and for agentic systems it has not been established what those records must contain for post-hoc causal attribution to be possible. We give the estimator framework and then the conditions under which it fails. We separate the marginal total effect that prior work measures from a common-random-numbers total effect that isolates a step's own contribution, add the natural direct effect under a pinned downstream, and check the estimators against hand derivations. Both estimands then fail, in the same direction. Under the marginal estimand a causally inert step has the identical total effect to the decisive one on every run of our planted chain, an algebraic identity and not a coincidence at one draw. Under common random numbers the decisive step returns exactly zero on the runs where the executing step flips, about one in ten, while its direct effect there is 0.25 and it demonstrably acts; an exact zero does not certify that a step did nothing, and we put that here rather than in the limitations. We derive the coupling that keeps the direct effect estimable once contexts diverge, with a closed form for its degradation, and show that the mediated share on which a natural ranking is built is not a share under suppression: where the direct and mediated paths oppose, it exceeds one and ranks a suppressed component above a pure mediator. We publish the discrepancy experiment's pre-registration rather than a result, because the live pipeline it requires was not available in the study window. We contribute the traceability specification such a filing would need, against a gap the Act's calendar opens: Article 86's right to an explanation has applied since 2 August 2026, while the Article 12 logging and Annex IV documentation that could evidence one were deferred to 2 December 2027 by Regulation (EU) 2026/1744.

The generation of safety-critical traffic scenarios is essential for training and evaluating autonomous vehicles. Prior approaches typically perturb the trajectories of existing agents in a traffic scenario using simplified adversarial objectives to induce safety-critical interactions, which can limit the plausibility and diversity of the generated scenarios. Although inserting new adversarial vehicles can alleviate this limitation, determining when and where to introduce them in a scenario-specific manner remains challenging. In this work, we introduce \underline{CO}llision \underline{S}napshot guided \underline{T}im\underline{E}-\underline{R}eversed safety-critical scenario generation (COSTER), a framework that leverages learned traffic priors to determine plausible collision times and locations. COSTER first constructs a collision snapshot by inserting a new vehicle in contact with the target vehicle at the identified collision state within a traffic scenario. Starting from this collision snapshot, a conditional variational autoencoder is used to perform a time-reversed rollout, reconstructing the trajectory of the inserted vehicle backward toward earlier timesteps. Experiments show that COSTER outperforms existing methods in plausibility, diversity, and data efficiency. Moreover, agents trained on COSTER-generated scenarios reduce collision rates by 31% on safety-critical scenarios from the Waymo Open Motion Dataset while also improving ego task completion. The project website is available at https://anonym-121.github.io/COSTER/.

Batch normalization (BN) substantially improves sample efficiency in continuous-control actor-critic methods such as CrossQ, yet recent studies report performance degradation in discrete-action value learning on Atari. These failures are surprising because discrete Q-networks lack the action-input distribution mismatch identified by CrossQ. We show for target-based C51 and target-free PQN that the simple choice between running and batch statistics at specific forward passes can reverse this degradation. In C51, switching the BN bootstrap forward to batch-statistic mode significantly improves performance over unnormalized and LayerNorm baselines and scales stably with update-to-data ratios up to 12. In PQN, using batch-statistics for both action selection and bootstrapping recovers performance from the failing running-statistic configuration. Across 26 Atari games at 400M frames, this configuration achieves a higher final aggregate score than PQN with LayerNorm. Our results show that carefully configured BN can substantially improve discrete-action value learning, and that its forward protocols are an essential part of the algorithm specification.

Test-time scaling produces cohorts of reasoning rollouts, yet there is no standard label-free account of how their internal computation reorganizes as inference unfolds. We introduce routing effective rank deff, the entropy-effective dimensionality of a cross-rollout graph built from MoE expert-routing similarity. Across ten MoE configurations and five math/science benchmarks, deff exhibits a reproducible low-high-low trajectory, with a prominent interior maximum in 98.5% of 3,105 model-question cohorts: routing similarity is concentrated early, maximally differentiated at intermediate budgets, and reconcentrated later, and the timing of this maximum varies systematically with architecture and reasoning effort. An exact decomposition separates cohort-wide common-mode mass from residual spectral dimensionality: common-mode reallocation accounts for about two thirds of the trajectory, while the residual spectrum contributes about one quarter and retains substantial variation beyond the common mode. The decomposition further localizes behavior: among non-unanimous cohorts, increases in common-mode concentration strongly predict same-answer recoverability, and higher reasoning effort delays the maximum by 2.59 octaves (doublings of the token budget) and consistently expands the high-rank period across all four tested architectures, locating the effort effect in timing and duration rather than peak amplitude. Correctness comparisons separate structural monitoring from answer selection, positioning routing effective rank as a decomposable, label-free diagnostic of cohort organization - a principled spectral lens on how MoE reasoning cohorts differentiate and reconcentrate over inference time.

Graph-agentic retrieval-augmented generation combines structured evidence with adaptive controllers that can plan retrieval, traverse relations, verify intermediate claims, delegate subtasks, and use tools. This combination is useful when answers depend on relations across documents, entities, time, or institutions, but it also creates coupled failure paths: a defect in graph construction can become retrieved evidence, alter later control decisions, and propagate toward a consequential outcome. We examine how such systems should be designed and evaluated for social-good settings in which freshness, authorization, traceability, oversight, and recourse matter alongside answer quality. We organize the literature by graph substrate, graph lifecycle, agent function, coordination pattern, and authority boundary, and distinguish graph-based retrieval from observation-dependent graph control. We then synthesize reported risks as an evidence-to-action failure chain and propose an assurance-by-construction blueprint comprising five interface contracts for evidence, retrieval, reasoning, capability and delegation, and outcome. These contracts make provenance, temporal validity, authorization, uncertainty, and recoverability explicit at system boundaries. An illustrative public-benefit information design shows how the framework constrains graph structure, permissions, abstention, and operating authority. Finally, we derive an evaluation agenda spanning graph assertions, trajectories, claims, coordination, and outcomes.

Motivated by the growing importance of fairness in machine learning, fair $k$-center clustering has attracted considerable research attention as a fundamental problem. In this problem, a dataset is partitioned into $m$ disjoint groups, and the objective is to select $k$ data points as centers, subject to upper bounds on the number of centers chosen from each group, aiming to minimize the maximum distance between any data point and its assigned center. Focusing on Euclidean spaces, which are ubiquitous in machine learning applications, we first develop a parameterized approximation algorithm for Euclidean fair $k$-center with an approximation ratio of $2.732$. By incorporating this algorithm as a post-processing stage into a one-pass streaming framework for large-scale data, we obtain an approximation ratio of $4.464$. These ratios can be further respectively improved to $2.414$ and $3.828$ with a runtime exponential on $k$. To ensure polynomial-time complexity, we further design a one-pass streaming algorithm with an approximation ratio of $4.732$, which can be further improved to $4.42$, outperforming the state-of-the-art ratio. Finally, extensive experiments show that our methods significantly outperform state-of-the-art approaches in terms of clustering accuracy.

In many scientific disciplines, weak signals of interest are obscured by dominant nuisance signals that are several orders of magnitude stronger. Recovering these weak signals requires subtracting the dominant ones; however, this calibration process inherently distorts or partially suppresses the underlying signal of interest. To address this problem, we propose the use of normalizing flow models to reconstruct calibration-affected weak signals. By leveraging the statistical invariance of the target signals and assuming minimal initial suppression, our framework effectively recovers the lost signal components. We provide a comprehensive theoretical overview of this normalizing flow-based recovery method and demonstrate its efficacy using simulated data.

LLM-as-a-Judge has emerged as a promising paradigm for evaluating natural language generation. However, the uncertainty associated with such evaluations remains largely unexplored, which limits their reliability in real-world applications. Although conformal prediction offers a principled framework for uncertainty quantification, existing approaches typically apply it to a single LLM judge, overlooking the variability introduced by using different LLM evaluators. In this work, we propose a robust uncertainty estimation framework for multi-agent LLM-as-a-Judge evaluation. Our approach constructs conformal prediction intervals for LLM-based scores from multiple LLMs. By considering intervals from different LLM judges, we obtain more stable and reliable uncertainty estimates. Extensive experiments demonstrate that our method produces valid prediction intervals with coverage guarantees, and that interval-based aggregation across multiple judges leads to more stable evaluation outcomes.

Scientific discovery in data-rich domains is currently constrained by human bandwidth: the growth in the volume and complexity of real-world data far outpaces the rate at which researchers can read, reason, and synthesize. Recent LLM-based multi-agent systems have begun to automate portions of the research cycle, but they target hypothesis generation in settings where validation cannot itself be automated, and each run is one-shot, with no mechanism for findings to accumulate or steer subsequent inquiry. This paper introduces AutoKD, a multi-agent framework for autonomous knowledge discovery that is both computational and cumulative, allowing validated findings to persist and inform subsequent inquiry. Six coordinated LLM agents collaborate in an open-ended discovery loop, where accepted findings are stored in a persistent insight graph that serves as both long-term memory and an exploration-steering mechanism. We evaluate AutoKD on three diverse datasets from two perspectives: Open-ended Quality against published findings, and Conditioned Quality via literature-derived queries. Across both evaluation perspectives, AutoKD covers known findings and surfaces substantive discoveries that complement human-driven research. Our code is available at https://github.com/GeQinwen/AutoKD.

Hyperspectral image (HSI) classification plays a vital role in remote sensing applications, including agriculture, environmental monitoring, and urban analysis. However, its performance remains challenged by high spectral redundancy, noise sensitivity, and the difficulty of jointly modeling local material composition and long-range spectral dependencies. To address this, we propose AGSA-Net, an abundance-guided self-attention network that explicitly integrates spectral unmixing priors into the classification process. AGSA Net first estimates physically meaningful subpixel abundance maps subject to non-negativity and sum-to-one constraints, regularized by hybrid linear-nonlinear reconstruction decoder. The learned abundances are then used to construct an abundance affinity prior that guides a spectral transformer to emphasize class-discriminative interactions, and the resulting transformer features are fused with compact abundance descriptors for final prediction; in contrast to existing approaches that use abundance as auxiliary or concatenated features. Experiments on Indian Pines, Augsburg, and Berlin demonstrate the benefit of incorporating abundance- guided contextual modeling, particularly in heterogeneous urban scenes. The source code and trained models are available at: https://github.com/nnuvi/AGSA-Net

Linear algebra provides the framework of concepts (matrix rank, singular value decomposition (SVD), and eigendecomposition) that modern artificial intelligence employs to encode, compress, and propagate information through neural networks. This paper unifies fourteen separate peer-reviewed works analyzing the usage of these techniques in the context of transformer-based foundation model research, focusing on three areas of the topic: derivations and properties of self-attention matrices' output rank, compression methods that purposefully utilize this phenomenon, and the low-rank key-value (KV) cache projection and its semiseparable-matrix duality to linear attention and state-space structured models. We were motivated to conduct this work after observing an open problem in this literature: the interplay of the mentioned compression methods with natural rank collapse of the network. With this paper, we report an original finding that using SVD compression of attention projections actually has the opposite effect on the rank collapse of the network: while it strongly suppresses it at initialization, it accelerates on pretrained models (for GPT-2 124M, GPT-2 Medium 355M, and Pythia-160M) with minimal risk of object aliasing artifacts appearing (verified on all compression ratios) and is consistent across four rank estimation methods. A controlled causal decomposition of the effect in both settings showed that the reason for this behavior can be explained by the choice of the subspace SVD makes when compressing the matrix better than the reduction of the operator norm it achieves, explaining roughly 76% of the effect at initialization and 83% on the pretrained weights, providing a refinement to the calibration-aware compression viewpoint and an explanation of why it outperformed naive SVD truncation.

The purpose of this research is to find data and methods using machine learning and deep learning to correctly predict the estimated travel time for transportation and logistics in a supply chain system. The supply chain ecosystem is very complex and heavily relies on the transportation and logistics of raw materials and finished goods. Accurate travel time estimation is critical because it helps supply chain members to improve logistics consistency and performance. This helps in planning, demand forecasting, lead time management and assembly planning. The logistics on the delivery side of the customer also plays a crucial role in customer satisfaction and voice of customer. With the collection of huge historical data and using novel techniques, the research builds an accurate model to predict travel time of inventory.

This paper treats prompt engineering as a discipline for turning informal human intent into structured AI work specifications. It develops the practice as a sequence of reusable design moves: define the work, construct only the context the answer depends on, choose a role, or a moderated panel of roles, as an attention lens, and state affirmative quality targets, reserving prohibitions for hard boundaries. To keep prompts lean, it adapts two classical principles, Occam's razor and Chekhov's gun, so that every instruction earns its place. For consequential tasks, it adds structured critique through steelmanning and premortems, followed by verification and, where tools or multi-step actions are involved, agentic operating loops with explicit boundaries and escalation. Aimed at a general readership, this tutorial is not a benchmarking study; it offers a practical, technically grounded path from casual prompting to disciplined AI workflow design, illustrated by a worked example that carries one task from a weak prompt to a strong specification. The framework is presented as principles and checklists that remain useful as models and tools change. The strongest prompt is rarely the longest prompt; it is the one that makes desired behaviour, required sources and checks, and success criteria unmistakable.

Systematic Literature Reviews (SLRs) are essential for evidence-based research but remain time-consuming, requiring researchers to manage large volumes of publications across planning, screening, analysis, and reporting. Large language models (LLMs) can now produce fluent, well-structured review text, which makes it difficult to distinguish synthesis that was verified by a researcher from synthesis that merely appears authoritative. This raises the risk that unverified AI-generated synthesis enters the scholarly record carrying the credibility of a systematic review. We present CoSLR, a Human-AI collaborative multi-agent system that supports the SLR workflow through a modular three-phase pipeline using large language models and Retrieval-Augmented Generation (RAG), and that places explicit, mandatory human checkpoints on the path between generated output and its acceptance. In a survey-based study with 63 participants, the system was received positively: 27 of 63 participants (42.9 percent) rated its usability highly, indicating that the mandatory checkpoints did not come at the cost of a workable interface. However, a checkpoint safeguards the review only if researchers use it to verify: 22 of 63 participants (34.9 percent) reported that they would trust AI-generated summaries and reports without additional human checking after only a short interaction with the system. These findings indicate that Human-AI collaboration can support literature review work, but that the effectiveness of human oversight depends on whether users are willing to exercise it. This is a calibration problem that interface design must address directly, not assume.

Search agents are usually trained under a single harness. But once an agent is deployed in a real application, its harness is frequently updated (e.g., a rewritten system prompt) to fit production needs. This exposes a fragility of post-trained agents: because a learned behavior is entangled with its training harness, even a harness update that leaves the task unchanged can fail to elicit the behavior. We train a search agent to perform parallel search, a popular strategy for improving both search efficiency and performance. We find that training under a fixed harness makes the behavior harness-local, overfit to that harness's surface form: when the harness changes, the model falls back to serial search. An intuitive fix is harness augmentation, but simply training on more harnesses does not resolve the problem. GRPO learns from the reward gap between parallel and serial rollouts of the same question: a small harness pool saturates that gap early, while a large pool dilutes the per-harness signal too thinly for any harness to consolidate. We therefore propose Curriculum HArness Rotation Training (CHART), a rotating curriculum that lets a search agent gradually consolidate parallel search across harnesses. At each periodic evaluation, CHART "graduates" the harnesses whose expected behavior is learned and replaces them with still-learnable ones, keeping the reward gap alive throughout training. Starting from the same harness pool, CHART makes the model learn parallel search on all harnesses, whereas static augmentation succeeds on at most half of them. The behavior also carries to held-out harnesses: CHART parallelizes on 89% of held-out turns, against at most 5% for the static pools. It further transfers to a new QA task and search environment, improving pass@1 by 5.6pp over the best static pool. Finally, CHART-trained agents benefit more from meta-harness search than baselines.

Large language models (LLMs) are increasingly used to forecast real-world events by retrieving and reasoning over news. We show that this dependence on an open, crawlable news corpus creates a new attack surface: an adversary who can merely publish articles–without access to the retriever, the model, or the user's queries–can systematically move the forecaster's output probabilities. We formalize news-corpus poisoning of probabilistic forecasters, a threat model distinct from prior RAG poisoning, which targets factual answers or opinion polarity rather than calibrated probabilities. We evaluate the attack on 500 resolved ForecastBench questions against a 17.4M-article Common Crawl News corpus with a strict crawl-date cutoff, using three retrieval-augmented forecasters built on open 7-8B models. A single LLM-written article per question flips 56% of forecasts across the 0.5 boundary; five articles flip 69-73% and shift probabilities by +0.13 to +0.22 net of a neutral-article placebo, degrading the Brier score from 0.18 to 0.37. The effect is monotone in the number, retrieval rank, query similarity, and context share of injected articles, transfers across model families, and is unaffected by the claimed publisher. We then evaluate three natural defenses–source allow-lists, isolate-then-aggregate forecasting, and perplexity filtering–and show that each has a cheap bypass: spoofed publishers, majority poisoning, and higher-temperature generation, respectively. Our results indicate that probabilistic LLM judgments inherit the full fragility of the information supply chain they consume.

Large language models can produce fluent explanations for chess moves, but plausible language does not necessarily reflect the reasoning behind a decision. We study this question in chess, where the board state is fully observable, legal actions can be enumerated, and move quality can be evaluated independently. Across 200 Lichess endgame puzzles, we test explanations using move recoverability, decoder-side controls, and token-level scoring of legal candidate moves. Unmasked explanations make generated moves easy to recover, but this advantage drops sharply after explicit move hints are removed. Under strict masking, explanations provide only small and decoder-dependent gains over the board state alone. Token-level scoring shows that explanations can nevertheless alter move preferences: random but plausible explanations from other puzzles reduce the probability of the correct move, indicating that irrelevant reasoning text is not simply ignored. We also find that recognizable endgame motifs can make generated moves easier to recover without reliably improving move correctness. Together, these results show that linguistic plausibility, consistency with a generated action, and solution correctness are distinct properties. Fluent chess explanations can influence action preferences and support a coherent move narrative while providing only limited evidence of faithful reasoning.

This research proposes the Universal Observatory Graph (UOG), an AI-driven framework for distributed astronomical observation across the Solar System. The proposed architecture models autonomous observatories located at the Sun planet L2 Lagrange points as nodes in a weighted graph, while communication links are represented as graph edges characterized by multi-objective physical and operational metrics, including interplanetary distance, communication latency, transmission power, and link reliability. The resulting graph provides a unified mathematical representation of a cooperative interplanetary observatory network. This proposal examines a six-observatory Solar System configuration comprising Earth, Mars, Jupiter, Saturn, Uranus and Neptune. Instantaneous sky coverage is evaluated independently using a 200,000 direction Fibonacci sphere, a 2,000,000 direction fixed seed Monte Carlo calculation and deterministic spherical integration. All three methods yield complete network union coverage, approximately 0.43% complete six observatory intersection and approximately 24.96% mean pairwise Jaccard similarity under the adopted pointing model. Communication routing is subsequently formulated as a finite horizon Markov decision process and solved using tabular Q-learning. The reward balances node participation and a distance dependent reliability proxy against distance, light time latency and a distance squared transmission power proxy. The learned Earth-Saturn-Uranus-Neptune route is also the highest discounted return route among all 41 feasible simple paths under the four hop constraint. The framework provides a reproducible baseline for sequential coverage assessment and multi objective routing; time dependent ephemerides, mission specific visibility, calibrated link budgets and scalable graph policies remain future work.

Input-conditioned neural interventions raise a runtime question: what persists when one behavioral specification admits multiple actions whose validity depends on execution state? We introduce replay-gated neural execution, separating five objects: a persistent behavioral predicate, its state-indexed certified realization set, a transient action witness, a budget-limited finder, and execution authorization. Candidates undergo isolated FP32/BF16 replay of the frozen model; commitment additionally requires a valid run audit. Experiments on Qwen3-0.6B and SmolLM2-360M-Instruct establish distinct failure modes for these objects. Independent initializations yield distinct certified actions in all 24 tested fixed-state cells. Unchanged SmolLM2 witnesses remain certified in all 128 native states but only 66 of 384 off-diagonal transfers. All 767 archived Qwen witnesses replay successfully, yet a budget-limited finder misses one known-realizable cell in all three prespecified runs. Of 1,141 replay-submitted candidates, 174 fail item certification. A frozen three-tier cascade uses these boundaries to reject uncertified proposals and escalate audit-valid search misses. On 256 previously sealed Qwen Fresh requests, 221 first certify at the lowest-cost tier and all 256 receive audited authorization, with no observed bypass. Relative to frozen full search, the median singleton search-and-certification cost ratio is 0.1055 and P95 is 1.3485, including failed tiers. Within the studied behavioral family on two small models, these results support state-indexed, set-valued execution semantics: specifications persist, search proposes witnesses, and replay certification plus run audit grants execution authority.

Multi-agent reinforcement learning in networked populations is governed by the interaction between individual adaptation, local encounters, and changing environmental conditions. To study this interaction, we formulate a coupled learning–environment model in which agents update stateless $Q$-values on a fixed graph, while their population-average behavior drives an environmental variable that dynamically modifies the payoff matrix. Under a first-order mean-field closure, we derive a deterministic transport equation for the population distribution of $Q$-values and couple it with a projected discrete update for the environmental state. The resulting model is evaluated against finite-network Monte Carlo simulations on random regular, Erdős–Rényi, Barabási–Albert, and random geometric graphs. Across the tested parameter ranges, the mean-field system reproduces the main macroscopic cooperation and environmental trajectories, and the trajectory-level root-mean-square error generally decreases with population size and average degree. The analysis further shows that environmental feedback reshapes the learned action-value ordering, while reinforcing feedback can produce pronounced dependence on the initial learning bias and resource level. The environmental timescale also plays an important role: a rapid response can drive the resource state to a boundary before learning adapts, whereas a slower response preserves the interaction between behavioral learning and environmental recovery. These results provide a population-level description of coupled reinforcement learning and environmental dynamics and characterize the performance of the mean-field approximation within the tested network and parameter ranges.

We present a machine-checked Lean~4 formalization of Dong and Yang's classification of optimal finite-length $(n,4)$ binary block codes for binary symmetric channels. The formalization was developed mainly by feeding the paper's proofs to an AI tool. To establish correctness, the authors verified the main theorem statements in Lean and the accepted axioms. This note discusses the corrections and simplifications made to the AI-generated formalization, and records discrepancies found in the paper during the formalization. The Lean code is available at https://github.com/shhyang/n4code_lean.

Existing benchmarks for autonomous AI scientists evaluate only final outputs—generated code, hypotheses, or papers—yet discard the reasoning process by which those outputs were obtained. This makes it impossible to audit scientific methodology, diagnose failure modes, or distinguish systematic reasoning from fortunate guessing. We present OpenDiscoveryTrace, a public dataset of 558 complete AI scientific agent trajectories that captures how models reason, not just what they produce. Each trajectory records a structured 9-field-per-step trace—including thoughts, tool calls, observations, errors, revision triggers, and self-reported confidence—as models execute 124 scientific tasks spanning drug discovery, materials science, genomics, and scientific literature analysis. The dataset covers seven models: three frontier models (GPT-5.4, Claude Opus 4.6, and Gemini 3.1 Pro; 124 trajectories each, fully balanced across domains and difficulty levels) and four open-weight models (Qwen2.5-7B, Mistral-7B-v0.3, Phi-3.5-mini, and Qwen2.5-1.5B; 30 each), plus 60 live-retrieval variant trajectories. Pilot analysis on 363 LLM-judged trajectories reveals that process traces expose behavioral differences invisible to output-only evaluation: all three frontier models achieve comparable success rates (84–89%), yet Claude Opus 4.6 produces 30$\times$ more errors than GPT-5.4 (2.5 vs. 0.08 per trajectory, $p < 0.0001$, Cliff's $\delta = 0.613$), with qualitatively different error profiles—66.7% tool misuse for Claude versus 83.6% reasoning errors for GPT-5.4. We define five benchmark tasks with baselines from logistic regression, random forests, LSTMs, and Transformer models. The dataset, trace schema, agent harness, and benchmark definitions are publicly available under CC BY 4.0 to support research on process-level evaluation, scientific agent auditing, and AI governance.

As large language models (LLMs) are increasingly deployed in alignment-sensitive contexts, activation steering has emerged as a lightweight, inference-time alternative to fine-tuning methods (e.g., RLHF, DPO) for behavioral control. However, existing work typically validates steering on isolated behaviors, leaving it unclear whether steering vectors encode coherent semantic structure or merely exploit behavior-specific shortcuts. We investigate whether the latent geometry of LLM steering vectors reflects theory-specified structure in human values and morality. Using Schwartz's Theory of Basic Human Values as our primary fine-grained framework, we introduce a 26K-sample benchmark covering 20 human values and analyze distribution-driven methods (e.g., CAA, SphericalSteer, ODESteer) and behavior-centric approaches (e.g., COLD-Steer, BiPO) across diverse model families and sizes. We find that distribution-driven methods recover human value topologies aligned with theoretical predictions (Spearman $\rho$ up to 0.51, $p < 10^{-13}$). In contrast, behavior-centric methods achieve comparable steering performance but show little correlation with the expected value geometry. Geometric fidelity improves with model scale but drops after instruction tuning. Finally, better geometric alignment also leads to more human-consistent transfer across values: steering one value correctly lifts compatible values and suppresses opposing ones. Code and data are available at: https://github.com/DeepRCL/Steering_Geometry.

Some low-cost Internet of Things (IoT) sensor deployments lack device-level source authentication, leaving them vulnerable to impersonation or injected sensor readings. We present a lightweight approach to sensor impersonation detection in a small proof-of-concept study. We formulate detection as a sequence-prediction problem. A model with three LSTM layers and two fully connected layers is trained only on univariate temperature readings from a genuine sensor, and a window of readings is flagged when its mean absolute prediction error exceeds the mean genuine error by more than six standard deviations. The model is converted to TensorFlow Lite in three variants, non-quantized (554 KB), 16-bit weight quantized (298.5 KB), and 8-bit weight quantized (185 KB), targeting an Arduino Nano 33 BLE aggregator. The variants were evaluated with the TensorFlow Lite interpreter on the local server, since on-device execution of LSTM models was not yet supported by TensorFlow Lite for Microcontrollers at the time of the study. On a controlled testbed with the impostor sensor placed in a hotter outdoor location, the three variants reached detection accuracies of 99.980%, 99.972%, and 98.206%, and each flagged the change point when a test sequence switched from genuine to impostor data. Quantization made the model smaller but slower in our measurements. The genuine and impostor distributions were well separated, so these results show detection of a controlled distribution shift and should not be read as evidence of general device authentication.

Security Operations Centers (SOCs) process large amounts of tickets, most of which are low-interest events not worthy of further investigation. The repetitive nature of this task and similarity of the vast amounts of tickets make it a prime candidate for generative AI-based automation. We created and deployed an agentic AI companion utilizing large language models through fieldwork within a SOC for over one year. The design of the SOC AI companion was driven by researchers' participation and interactions within the SOC's daily work. SOC analysts were invited to use it during the last four months of the fieldwork. We analyzed the analysts' usage of the companion and found that in more than 90% of the cases the companion's outputs were reused by analysts in the ticket's closing report. Our results showed that when designed "in the trenches" with the intended users, a SOC AI companion can go beyond being yet another tool, but rather a system that co-evolves with its human users as it traverses through the various types of workloads. Analysts naturally started to shape the AI companion's behaviors to fit their particular needs. Our data show that the more human analysts shape the AI companion's behaviors, the more they become comfortable trusting the output from the AI system, resulting in improved productivity.

Multimodal Large Language Models (MLLMs) perform strongly on general visual understanding tasks such as visual question answering, yet they often struggle with a basic comparative skill: identifying what has changed between two similar images. We introduce VDiff-Bench, a challenging multiple-choice benchmark for fine-grained Image Difference Identification. VDiff-Bench contains 1,756 four-way questions over image pairs and covers 10 change categories: position, motion, regional image color, overall image color, appearance/disappearance, noise/resolution, texture, substitution/size, OCR/text, and illumination. Each question corresponds to two image inputs with 4 choices: the true difference, two hard negative descriptions, and a "no difference" distractor. To make the task challenging, we specifically curate ground-truth-conditioned negatives that require models to distinguish the actual change from nearby semantic alternatives. Experiments with 11 state-of-the-art open- and closed-source MLLMs show that fine-grained visual comparison remains brittle: models exhibit uneven performance across sources and change categories, with persistent failures on subtle low-level changes like noises and textures. For instance, three 7-8B-scale open-source MLLMs score 52.5-70.6% on semantic changes but only 8.7-33.3% on low-level changes like noise and texture, falsely assuming no changes between two image inputs. Surprisingly, despite strong performance of other closed-source commercial models, Grok 4.3 demonstrate remarkable performance drop on identifying noise and texture differences between images, falling significantly behind large open-source models like Kimi K2.5 and K3. Overall, VDiff-Bench provides a targeted diagnostic for evaluating comparative visual understanding in MLLMs, exposing failures that are not captured by standard single-image vision-language tasks.