arXiv subjects

Aaron Chan

Publications and source records attributed to Aaron Chan.

At least 19 recordsLinked to original sources

arXiv2026

FIDES: A Concordance Protocol for LLM-Generated Trading Strategies

An LLM asked for a trading strategy returns three artifacts at once: a natural-language rationale, an executable implementation, and once run, a track record. Whether these are the same object is rarely checked. We present FIDES, a measurement protocol that treats them as three views to be reconciled rather than one deliverable to be graded. Through dual delivery, a single model call returns both a natural-language strategy with an explicit claimed edge and a self-contained strategy(df) function. FIDES executes the code in a sandbox against a lag-one out-of-sample backtest and scores three concordance gaps: say to do, do to real, and say to result. On 8 liquid US ETFs across four models plus a two-stage elicitation arm, 40 strategies, 2023 to 2024 out-of-sample, three findings stand out. First, concordance does not predict profit: only 2 of 40 strategies beat buy-and-hold, and a plain sma(50,200) rule outperforms every model's mean Sharpe. Second, self-assessment is badly calibrated: 32 of 40 strategies claim to beat buy-and-hold and exactly one does. Third, swapping the language-code judge for a second model flips say to do on more than half of items. Injecting Close.shift(-1) drops do to real by 0.33 on average, while our runtime future-information probe fired on neither clean nor injected code. We frame FIDES as a protocol for measurement fidelity, not a claim about market performance.

cs.CR↗

arXiv2026

SFGA: A Statistics-First Gating Architecture with Adjudicative Escalation for Trustworthy SFT Data Procurement

Procuring supervised fine-tuning (SFT) data forces a buyer to decide, before any downstream training, whether a candidate corpus is worth acquiring. We present \sys{}, a statistics-first gating architecture that treats procurement as a cost-aware routing problem over three intrinsic quality axes -- diversity, utility, and redundancy. Cheap blind measurements are summarised into per-axis estimates with confidence intervals; a gate accepts a decision only when intervals are tight, sample sizes are adequate, and the axes agree, otherwise it escalates the case to an adjudicative debate between a buy-advocate and a reject-advocate judge, resolved by a presiding verdict. On a controlled benchmark of 12 datasets ($2{\times}3{\times}2$ grid over the three axes) with 5 seeds, the gate reaches 0.90 accuracy and 0.83 $F_1$ at \$0.017 per unit, sitting between an always-verify baseline (0.75) and an oracle upper bound (0.98) while spending less than always-escalate (\$0.020). We further report honest negative diagnostics of the debate path: a con-side win rate of 0.80 ($p\approx3{\times}10^{-6}$) and a 52\% position-flip rate under advocate swapping expose negativity and positional biases that a naive LLM-judge would hide. We frame the injected-knob evaluation explicitly as a controlled synthetic benchmark for measurement fidelity and routing calibration, and delimit external validity as future work.

cs.LG↗

arXiv2026

Preprojective algebras and generalisations: A short survey

The preprojective algebra of a hereditary algebra $H$ can be defined as a certain orbit construction of the regular representation generated by the Auslander-Reiten translation. In this short survey, we will look at two important generalisations, namely, the contracted preprojective algebra and the total preprojective algebra. We will include several open problems and questions motivated by examples in the hope to stimulate future research on general orbit algebras of $H$-modules.

math.RT↗

arXiv2026

LATTICE: Evaluating Decision Support Utility of Crypto Agents

We introduce LATTICE, a benchmark for evaluating the decision support utility of crypto agents in realistic user-facing scenarios. Prior crypto agent benchmarks mainly focus on reasoning-based or outcome-based evaluation, but do not assess agents' ability to assist user decision-making. LATTICE addresses this gap by: (1) defining six evaluation dimensions that capture key decision support properties; (2) proposing 16 task types that span the end-to-end crypto copilot workflow; and (3) using LLM judges to automatically score agent outputs based on these dimensions and tasks. Crucially, the dimensions and tasks are designed to be evaluable at scale using LLM judges, without relying on ground truth from expert annotators or external data sources. In lieu of these dependencies, LATTICE's LLM judge rubrics can be continually audited and updated given new dimensions, tasks, criteria, and human feedback, thus promoting reliable and extensible evaluation. While other benchmarks often compare foundation models sharing a generic agent framework, we use LATTICE to assess production-level agents used in actual crypto copilot products, reflecting the importance of orchestration and UI/UX design in determining agent quality. In this paper, we evaluate six real-world crypto copilots on 1,200 diverse queries and report breakdowns across dimensions, tasks, and query categories. Our experiments show that most of the tested copilots achieve comparable aggregate scores, but differ more significantly on dimension-level and task-level performance. This pattern suggests meaningful trade-offs in decision support quality: users with different priorities may be better served by different copilots than the aggregate rankings alone would indicate. To support reproducible research, we open-source all LATTICE code and data used in this paper.

cs.CR↗

arXiv2026

Fractionally Calabi-Yau algebras and cluster tilting

We show that the class of twisted fractionally Calabi-Yau algebras of finite global dimension coincides with the stable endomorphism algebras of $d$-cluster tilting modules over $d$-representation-finite algebras. This is an application of our main result stating that an algebra $A$ of finite global dimension is twisted fractionally Calabi-Yau if and only if there exists $i$ such that the replicated algebra $A^{(i)}$ is a higher Auslander algebra if and only if there exist infinitely many $i$ such that $A^{(i)}$ is a higher Auslander algebra. This gives a new connection between the study of higher Auslander-Reiten theory and twisted fractionally Calabi-Yau algebras, and provides a new construction of large classes of higher Auslander algebras and higher representation-finite algebras. We give several applications such as an explicit characterisation of twisted $\frac{n}{2}$-Calabi-Yau algebras, and a triangle equivalence between the bounded derived category of a twisted fractionally Calabi-Yau algebra of finite global dimension and the $\mathbb{Z}$-graded stable module category of an associated higher preprojective algebra.

math.RT↗

arXiv2026

PoQ-Judge: A Multi-Architecture Evaluation Framework for Cost-Aware Proof-of-Quality in Decentralized LLM Inference

Decentralized LLM inference networks need lightweight, reference-free quality evaluation for Proof of Quality (PoQ). We present PoQ-Judge, a framework that trains dedicated judge models to score query-output pairs without ground-truth references. We study three architectures across the quality-cost tradeoff: a TextCNN judge, a MiniLM cross-encoder, and a DeBERTa judge. Using two-stage training on UltraFeedback plus GPT-labeled in-domain data, the best model reaches 0.747 Pearson correlation with the ground-truth proxy on a held-out test set, outperforming reference-based evaluators from prior work. As a reference-free component in composite scoring, it achieves 0.645 Pearson correlation, matching the best single reference-based evaluator while removing the need for reference answers. We also show that online calibration identifies semantic quality as the dominant dimension and that cascade evaluation reduces cost by 72.7 percent with only modest quality loss. Results are much stronger on QA than summarization, pointing to proxy quality as the main remaining limitation.

cs.CL↗

arXiv2026

Composer 2 Technical Report

Composer 2 is a specialized model designed for agentic software engineering. The model demonstrates strong long-term planning and coding intelligence while maintaining the ability to efficiently solve problems for interactive use. The model is trained in two phases: first, continued pretraining to improve the model's knowledge and latent coding ability, followed by large-scale reinforcement learning to improve end-to-end coding performance through stronger reasoning, accurate multi-step execution, and coherence on long-horizon realistic coding problems. We develop infrastructure to support training in the same Cursor harness that is used by the deployed model, with equivalent tools and structure, and use environments that match real problems closely. To measure the ability of the model on increasingly difficult tasks, we introduce a benchmark derived from real software engineering problems in large codebases including our own. Composer 2 is a frontier-level coding model and demonstrates a process for training strong domain-specialized models. On our CursorBench evaluations the model achieves a major improvement in accuracy compared to previous Composer models (61.3). On public benchmarks the model scores 61.7 on Terminal-Bench and 73.7 on SWE-bench Multilingual in our harness, comparable to state-of-the-art systems.

cs.SE↗

arXiv2026

Proof-of-Guardrail in AI Agents and What (Not) to Trust from It

As AI agents become widely deployed as online services, users often rely on an agent developer's claim about how safety is enforced, which introduces a threat where safety measures are falsely advertised. To address the threat, we propose proof-of-guardrail, a system that enables developers to provide cryptographic proof that a response is generated after a specific open-source guardrail. To generate proof, the developer runs the agent and guardrail inside a Trusted Execution Environment (TEE), which produces a TEE-signed attestation of guardrail code execution verifiable by any user offline. We implement proof-of-guardrail for OpenClaw agents and evaluate latency overhead and deployment cost. Proof-of-guardrail ensures integrity of guardrail execution while keeping the developer's agent private, but we also highlight a risk of deception about safety, for example, when malicious developers actively jailbreak the guardrail. Code and demo video: https://github.com/SaharaLabsAI/Verifiable-ClawGuard

cs.CR↗

arXiv2026

A Multi-Dimensional Quality Scoring Framework for Decentralized LLM Inference with Proof of Quality

Decentralized large language model (LLM) inference networks can pool heterogeneous compute to scale serving, but they require lightweight and incentive-compatible mechanisms to assess output quality. Prior work introduced cost-aware Proof of Quality (PoQ) and adaptive robust PoQ to allocate rewards under evaluator heterogeneity and adversarial behavior. In this paper, we focus on the quality signal itself and propose a multi-dimensional quality scoring framework that decomposes output quality into modular dimensions, including model and cost priors, structure quality, semantic quality, query-output alignment, and agreement/uncertainty. Using logged outputs from QA and summarization tasks, we systematically audit dimension reliability and show that seemingly reasonable dimensions can be task-dependent and even negatively correlated with reference quality without calibration. While the default composite underperforms a strong single semantic evaluator, ablations reveal that removing unreliable dimensions and re-normalizing weights yields a calibrated composite that matches or exceeds the best single- evaluator and consensus baselines. Finally, we integrate the composite score as a drop-in quality signal in PoQ and demonstrate complementary benefits with robust aggregation and adaptive trust weighting under adversarial evaluator attacks.

cs.LG↗

arXiv2026

Adaptive and Robust Cost-Aware Proof of Quality for Decentralized LLM Inference Networks

Decentralized large language model inference networks require lightweight mechanisms to reward high quality outputs under heterogeneous latency and cost. Proof of Quality provides scalable verification by sampling evaluator nodes that score candidate outputs, then aggregating their scores into a consensus signal that determines rewards. However, evaluator heterogeneity and malicious score manipulation can distort consensus and inflate payouts, which weakens incentive alignment in open participation settings. This paper extends a cost-aware Proof of Quality mechanism by adding adversary-resilient consensus formation. We study robust aggregation rules, including median and trimmed mean, and an adaptive trust-weighted consensus that updates evaluator weights from deviation signals. Using question answering and summarization workloads with a ground truth proxy for offline analysis, we quantify evaluator reliability and show strong variance across evaluators, including task-dependent misalignment that can invert correlations. We then evaluate robustness under four adversarial strategies, including noise injection, boosting, sabotage, and intermittent manipulation, across a sweep of malicious ratios and evaluator sample sizes. Our results show that robust aggregation improves consensus alignment with the ground truth proxy and reduces sensitivity to noisy and strategic attacks compared with simple averaging. We further characterize the operational trade-off introduced by evaluator sampling, where larger evaluator sets reduce evaluator rewards and increase payoff variance while inference rewards remain relatively stable in our configuration. These findings motivate robust consensus as a default component for cost-aware Proof of Quality and provide practical guidance for selecting evaluator sampling parameters under adversarial risk and resource constraints.

cs.CR↗

arXiv2025

Optimistic TEE-Rollups: A Hybrid Architecture for Scalable and Verifiable Generative AI Inference on Blockchain

The rapid integration of Large Language Models (LLMs) into decentralized physical infrastructure networks (DePIN) is currently bottlenecked by the Verifiability Trilemma, which posits that a decentralized inference system cannot simultaneously achieve high computational integrity, low latency, and low cost. Existing cryptographic solutions, such as Zero-Knowledge Machine Learning (ZKML), suffer from superlinear proving overheads (O(k NlogN)) that render them infeasible for billionparameter models. Conversely, optimistic approaches (opML) impose prohibitive dispute windows, preventing real-time interactivity, while recent "Proof of Quality" (PoQ) paradigms sacrifice cryptographic integrity for subjective semantic evaluation, leaving networks vulnerable to model downgrade attacks and reward hacking. In this paper, we introduce Optimistic TEE-Rollups (OTR), a hybrid verification protocol that harmonizes these constraints. OTR leverages NVIDIA H100 Confidential Computing Trusted Execution Environments (TEEs) to provide sub-second Provisional Finality, underpinned by an optimistic fraud-proof mechanism and stochastic Zero-Knowledge spot-checks to mitigate hardware side-channel risks. We formally define Proof of Efficient Attribution (PoEA), a consensus mechanism that cryptographically binds execution traces to hardware attestations, thereby guaranteeing model authenticity. Extensive simulations demonstrate that OTR achieves 99% of the throughput of centralized baselines with a marginal cost overhead of $0.07 per query, maintaining Byzantine fault tolerance against rational adversaries even in the presence of transient hardware vulnerabilities.

cs.CR↗

arXiv2025

Design and Evaluation of Cost-Aware PoQ for Decentralized LLM Inference

Decentralized large language model (LLM) inference promises transparent and censorship resistant access to advanced AI, yet existing verification approaches struggle to scale to modern models. Proof of Quality (PoQ) replaces cryptographic verification of computation with consensus over output quality, but the original formulation ignores heterogeneous computational costs across inference and evaluator nodes. This paper introduces a cost-aware PoQ framework that integrates explicit efficiency measurements into the reward mechanism for both types of nodes. The design combines ground truth token level F1, lightweight learned evaluators, and GPT based judgments within a unified evaluation pipeline, and adopts a linear reward function that balances normalized quality and cost. Experiments on extractive question answering and abstractive summarization use five instruction tuned LLMs ranging from TinyLlama-1.1B to Llama-3.2-3B and three evaluation models spanning cross encoder and bi encoder architectures. Results show that a semantic textual similarity bi encoder achieves much higher correlation with both ground truth and GPT scores than cross encoders, indicating that evaluator architecture is a critical design choice for PoQ. Quality-cost analysis further reveals that the largest models in the pool are also the most efficient in terms of quality per unit latency. Monte Carlo simulations over 5\,000 PoQ rounds demonstrate that the cost-aware reward scheme consistently assigns higher average rewards to high quality low cost inference models and to efficient evaluators, while penalizing slow low quality nodes. These findings suggest that cost-aware PoQ provides a practical foundation for economically sustainable decentralized LLM inference.

cs.AI↗

arXiv2025

Thermoelectric evidence of the electronic structure changes from the charge-density-wave transition in FeGe

Kagome metals provide a material platform for probing new correlated quantum phenomena due to the naturally incorporated linear dispersions, flat bands, and Van Hove singularities in their electronic structures. Among these quantum phenomena is the charge density wave (CDW), or the distortion of the lattice structure due to the motion of correlated electrons through the material. CDWs lower the energy of the compound, creating an energy gap that facilitates behaviors akin to superconductivity, nonlinear transport, or other quantum correlated phenomena. The kagome metal FeGe has been shown to host a CDW transition at approximately 100 K, and its occurrence is strongly influenced by the sample annealing conditions. However, a notable gap in the literature is the lack of clear thermoelectric transport evidence for electronic structure changes associated with this CDW transition. Here we present evidence of electron behavior modification due to annealing disorder via thermoelectric measurements on FeGe crystals presenting a CDW transition and those without a CDW. The observed Nernst effect and Seebeck effect under sufficient annealing demonstrate modified electrical transport properties resulting from induced disorder, including a change in carrier sign and an enhancement of the Nernst effect due to the CDW. Our results provide evidence of multiple phase transitions, which confirms the influence of CDW on the thermal properties of FeGe and demonstrates the suppression of CDW with sufficient disordering.

cond-mat.str-el↗

arXiv2025

Tilting theoretic approach to quasi-hereditary structures

A quasi-hereditary algebra is an algebra equipped with a certain partial order $\unlhd$ on its simple modules. Such a partial order -- called a quasi-hereditary structure -- gives rise to a characteristic tilting module $T_{\unlhd}$ by a classical result due to Ringel. A fundamental question is to determine which tilting modules can be realised as characteristic tilting modules. We answer this question by using the notion of IS-tilting module, which is a pair $(T,\unlhd)$ of a tilting module $T$ and a partial order $\unlhd$ on its direct summands such that iterative idempotent truncation along $\unlhd$ always reveals a simple direct summand. Specifically, we show that a tilting module $T$ is characteristic if, and only if, there is some $\unlhd$ so that $(T,\unlhd)$ is IS-tilting; in which case, we have $T=T_{\unlhd}$. This result enables us to study quasi-hereditary structures using tilting theory. As an application of the above result, we show that, for an algebra $A$, all tilting modules are characteristic if, and only if, $A$ is a quadratic linear Nakayama algebra. Furthermore, for such an $A$, we provide a decomposition of the set of its tilting modules that can be used to derive a recursive formula for enumerating its quasi-hereditary structures. Finally, we describe the quasi-hereditary structures of $A$ via `nodal gluing' and binary tree sequences.

math.RT↗

arXiv2025

SOP-Bench: Complex Industrial SOPs for Evaluating LLM Agents

LLM-based agents struggle to execute complex, multi-step Standard Operating Procedures (SOPs) that are fundamental to industrial automation. Existing benchmarks fail to capture the procedural complexity and tool orchestration demands of real-world workflows. We introduce SOP-Bench, a benchmark of 2,000+ tasks from human expert-authored SOPs across 12 business domains (healthcare, logistics, finance, content moderation, etc.). Using a human-AI collaborative framework, experts crafted authentic SOPs while AI generated artifacts (tools, APIs, datasets), all human-validated, yielding realistic tasks with executable interfaces and ground-truth outputs. SOP-Bench serves as a research enabler for systematically investigating agent architectures, model capabilities, and deployment considerations across diverse procedural tasks. We demonstrate its utility through illustrative experiments with a subset of frontier models across Function-Calling (FC) and ReAct agents, revealing critical insights. For example, (1) newer models do not guarantee better performance - Claude 4 family outperforms Claude 4.5 family on ReAct tasks (Claude 4 Opus: 72.4% vs. Claude 4.5 Sonnet: 63.3% task success rate), demonstrating that production upgrades require validation; (2) no single model-agent combination dominates: best performances range from 57% to 100% depending on domain. These examples illustrate how SOP-Bench enables isolating and studying specific dimensions of agent performance without costly production experiments. Our goal is not to rank model capabilities or build optimal agents, but to provide a rigorous evaluation framework that enables the researchers and practitioners to systematically investigate agent design choices, model selection, and deployment strategies. We release the benchmark at https://github.com/amazon-science/sop-bench.

cs.AI↗

arXiv2025

Quantum critical electro-optic and piezo-electric nonlinearities

Electro-optics, the tuning of optical properties of materials with electric fields, is key to a multitude of quantum and classical photonics applications. However, a major obstacle preventing many emerging use cases is inefficient modulation in cryogenic environments, as traditional tuning mechanisms degrade at low temperatures. Guided by the connection between phase transitions and nonlinearity, we identify the quantum paraelectric perovskite SrTiO$_3$ (STO) as the strongest cryogenic electro-optic photonic material. As a result of the unique quantum paraelectric phase of STO, we demonstrate a dynamically tunable linear Pockels coefficient ($r_{33}$) exceeding 500 pm/V at $T=5$ K, and study its full temperature and bias dependence. We also measure an enhanced piezo-electric coefficient ($d_{33}$) above 90 pC/N. Both of these coefficients exceed all previously reported values for cryogenic materials, including lithium niobate ($r_{33}\approx24$ pm/V) and barium titanate ($r_{42}\approx170$ pm/V). Furthermore, by tuning STO towards \textit{quantum criticality} with oxygen isotope substitution we more than double the optical and piezo-electric nonlinearities, demonstrating a linear Pockels coefficient above 1100 pm/V. Our results probe the link between quantum phase transitions, dielectric susceptibility, and optical nonlinearities, unlocking opportunities in cryogenic optical and mechanical systems, and provide a framework for discovering new nonlinear materials.

cond-mat.mtrl-sci↗

arXiv2025

Total preprojective algebras

We introduce total preprojective algebras $\Psi$ of path algebras of Dynkin quivers $kQ$, and prove that they are isomorphic to $2$-Auslander algebras of preprojective algebras $\Pi$ of $kQ$. In particular, $\Psi$ has global dimension $3$ and dominant dimension $3$. We also describe $\Psi$ as a tensor algebra of a certain explicit bimodule over the Auslander algebra of $kQ$. As an application, we give a presentation of $\Psi$ by explicit quivers with relations. More generally, we introduce total $(d+1)$-preprojective algebras of $d$-representation finite algebras, and give all the corresponding results.

math.RT↗

arXiv2025

Quantum oscillations in the heat capacity of Kondo insulator YbB12

We observe the magnetic quantum oscillation in the heat capacity of the Kondo insulator YbB$_{12}$. The frequency of these oscillations $F = 670$ T, aligns with findings from magnetoresistance and torque magnetometry experiments for $\mu_0 H > 35$ T in the Kondo insulating phase. Remarkably, the quantum oscillation amplitudes in the heat capacity are substantial, with $\Delta \tilde{C}/T \approx$ 0.5 $\rm{mJ}$ $\rm{mol^{-1}K^{-2}}$ at 0.8 K, accounting for 13$\%$ of the known linear heat capacity coefficient $\gamma$. Double-peak structures of quantum-oscillation amplitudes due to the distribution function of fermions were identified and used to determine the value of the effective mass from the heat capacity, which agrees well with that from torque magnetometry. These observations support charge-neutral fermions contributing to the quantum oscillations in YbB$_{12}$.

cond-mat.str-el↗