Searcharxiv⌕ Search

arXiv subjects

Wei Qiu

Publications and source records attributed to Wei Qiu.

At least 19 recordsLinked to original sources
arXiv2026

Increased throughput in antisymmetrically actuated acoustofluidic flow-through devices

Separation of low-abundance biological objects requires high throughput for practical use of an acoustofluidic system. Increasing the flow rate helps in achieving high-throughput if the acoustic energy density can be increased proportionally, and this may be possible with an efficient coupling of the transducer to the device. In particular, antisymmetric actuation using two electrodes with opposite phases is theoretically proven to enhance the acoustic energy density of the device. In this work, we study the symmetric and antisymmetric actuation mechanisms of an acoustofluidic system using both experiments and three-dimensional numerical simulations. The acoustic focusability experiments show that under the same electrical input power, the antisymmetric actuation mode performs better than the symmetric actuation, quantified in terms of the normalized width of the band formed by the focused particles. Numerical simulations of this particle bandwidth are performed for both actuation modes, and the results suggest that the antisymmetric actuation mode is more robust than the symmetric one, being weakly dependent of the geometric symmetry properties of the system. The simulation results corroborate the experimental findings, which indicate that the antisymmetric actuation increases the acoustophoretic efficiency and robustness for high-throughput applications.

physics.flu-dyn↗
arXiv2026

ECG-InterpBench: Benchmarking the Interpretability of ECG Foundation Models with Matched-Scale Sparse Autoencoders

Existing benchmarks for electrocardiogram foundation models primarily evaluate downstream predictive performance, providing limited insight into whether their internal representations can be faithfully decomposed, clinically interpreted, or reproduced across independent analyses. We introduce ECG-InterpBench, a benchmark designed to systematically evaluate the interpretability of ECG foundation-model representations. ECG-InterpBench uses sparse autoencoders as standardized measurement instruments and matches their capacity across models to enable controlled comparisons. We evaluate six frozen ECG foundation models across five standardized encoder depths, five matched dictionary widths, and three random seeds, producing a 450-cell interpretability atlas comprising 75 exactly matched six-model comparison blocks. The benchmark evaluates complementary dimensions of representation interpretability, including sparse reconstruction fidelity, single-feature accessibility and coverage of 49 clinically meaningful ECG measurements, and cross-seed feature reproducibility. The evaluation further quantifies patient-sampling uncertainty, depth- and seed-dependent variation, and sensitivity to the sparsity parameterization. The benchmark reveals that ECG foundation models exhibit distinct interpretability profiles. A matched replication on MIMIC-IV-ECG confirms that reconstruction fidelity and clinical accessibility identify different leading models. The benchmark is accompanied by executable evaluation code, standardized manifests, cell-level metrics, and reproducibility audits. ECG-InterpBench complements performance-centered ECG benchmarks by providing a capacity-controlled and reproducible framework for comparing ECG foundation models across distinct dimensions of representation interpretability.

cs.LG↗
arXiv2026

OrganLens: Organ-Specific Representation Learning for CT Foundation Models

A CT examination captures multiple organs, but many biomedical questions concern abnormalities, prognosis, or longitudinal change in a specific organ. These questions require a separate representation for each organ within the same CT volume. Existing CT foundation models commonly produce a single volume-level representation, while recent anatomy-aware methods either encode pre-separated organ volumes or explicitly disentangle images into organ token groups. The former may remove clinically relevant surrounding context, while the latter does not condition a shared encoder on a selected organ before its features are formed. We introduce OrganLens for organ-specific representation learning through self-supervision. An organ identity conditions a shared CT encoder, while organ-specific distillation and anatomy-mask supervision shape features for anatomy-weighted pooling into organ-specific representations. At inference, the shared model produces 11 organ-specific representations without external segmentation masks. We evaluate OrganLens on CT-RATE, RAD-ChestCT, INSPECT, and NLST across diverse acquisitions and downstream evaluations. Relative to CT-pretrained DINOv2, heart representations raise CT-RATE cardiomegaly AUROC from 0.910 to 0.953, while lung representations improve the Harrell C-index for NLST lung-cancer mortality by 14.2\%. The global representation reaches INSPECT Recall@10 of 33.09\% and 32.04\% for text-to-image and image-to-text retrieval, respectively. Across organ-related tasks, anatomically matched representations provide stronger task-relevant signal, while the global representation retains broad utility. OrganLens offers a scalable approach to organ-specific CT representation learning with a shared encoder. More broadly, it provides the medical research community with a reusable framework for studying organ-specific disease across cohorts and clinical endpoints.

cs.CV↗
arXiv2026

CADENCE: A Cardiac Atom Dictionary for Interpretable Neural Concept Extraction from ECG Foundation Models

Foundation models for 12-lead electrocardiograms (ECGs) transfer well across clinical tasks, but the physiological knowledge encoded in their representations remains opaque. We present CADENCE, a framework that decomposes an ECG foundation model into a human-interpretable, queryable dictionary of physiological concepts. Using a BatchTopK sparse autoencoder, CADENCE factorizes Layer-6 embeddings from more than nine million ECG tokens into 8,192 sparse cardiac atoms. These atoms align better than individual dense embedding dimensions with clinical phenotypes and waveform morphology, recovering arrhythmias, conduction abnormalities, infarction and repolarization patterns, chamber and axis findings, and lead- and beat-phase-specific waveform primitives. At Layer 6, the best atoms achieve mean AUROCs of 0.88 for clinical phenotypes and 0.90 for morphology, versus 0.78 and 0.83 for the best dense dimensions. Sparse atom probes match or outperform dense probes for phenotype, morphology, and age prediction while attributing each prediction to a small set of interpretable atoms; phenotype AUROC improves from 0.93 to 0.95. Atom-space geometry recovers physiologically coherent relationships, and targeted atom ablation selectively changes frozen downstream outputs. An automated LLM pipeline generates and quantitatively validates atom descriptions by predicting held-out activations. On independent external ECG datasets, CADENCE recovers overlapping concepts and maintains consistent phenotype-prediction performance. CADENCE provides a scalable framework for discovering and auditing the physiological knowledge encoded by ECG foundation models.

cs.AI↗
arXiv2026

PyroDash: Cost-Efficient Token-Level Small-Large Language Model Collaborative Inference

Large language models (LLMs) provide strong reasoning capabilities but are expensive to serve at scale, whereas small language models (SLMs) are cheaper but less reliable on difficult problems. We introduce PyroDash, a cost-aware framework for token-level SLM-LLM collaborative inference. During generation, the SLM decides whether to request assistance by emitting a control token. A Collaborate Engine then sends the query and partial reasoning trace to a frozen LLM for completion through a single handoff. The policy is internalized in the SLM, requiring neither a separate router, LLM retraining, nor access to LLM logits. PyroDash trains the SLM in three stages: control-token embedding learning, offloading-oriented supervised fine-tuning, and cost-aware alignment with Group Relative Policy Optimization. Its reward balances answer accuracy against inference cost normalized by LLM-only inference. Across five mathematical reasoning benchmarks, PyroDash supports different accuracy-cost operating points. With $\lambda=0.05$, it achieves 64.04 percent average accuracy, 6.36 percentage points above the LLM-only baseline, while reducing cost by 20.4 percent. With $\lambda=0.6$, it achieves 54.55 percent accuracy with a 1.90 percent LLM token ratio and 0.012 LLM calls per example, reducing total cost from USD 49.36 to USD 1.78. These results show that learned token-level handoffs can reduce LLM use while preserving strong reasoning performance.

cs.CL↗
arXiv2026

Who Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents

Self-evolving agent systems create, revise, and retire their own skills, but every such loop assumes a reliable evaluation metric already exists. In many real applications none does. We show the metric itself can be the evolving object: our loop searches compositions of small typed drawback detectors under a full evolutionary lifecycle, selecting for agreement with a ten-item anchored reference set and regularizing by consensus over unlabeled outputs. What evolves is the function that grades one output, never the fixed task sets it is scored on, and what comes out is an inspectable expression rather than an opaque judge. It is also valid: on code generation it gains 0.21 agreement with hidden ground truth on a locked set that metric selection never reads (paired $p=0.014$), beating the bare LLM judge it contains. Validity is where safety lives: removing the anchor guards collapses the metric into a vacuous always-pass detector while removing the detector lifecycle does not, inverting the lesson from skill evolution. That collapse warns this line of work that downstream task score cannot validate a self-evolved evaluator, since the collapsed metric trains skills just as well. Task score answers only sufficiency, and an evolved metric suffices: \emph{Double Ratchet}, co-evolving the metric with a lifecycle-managed skill loop, retains 88--110\% of the lift ground truth or a hand-written rubric buys, across MBPP+, Spider~2.0-Snow, and report generation. When evolved skills gamed the report rubric, an independent judge caught it and one added detector repaired it.

cs.AI↗
arXiv2026

The Blind Curator: How a Biased Judge Silently Disables Skill Retirement in Self-Evolving Agents

A self-evolving agent retires its bad skills by watching them fail, so what happens when the judge cannot see the failures? Skill retirement is the structural constraint that keeps a growing library from drifting below the no-skill baseline, but its guarantee assumes an unbiased reward, which is false for the LLM judges that reference-free tasks require. We show that a biased judge does not merely add noise; it \emph{silently switches off the curator}. We make this precise with a corrupted-reward analysis, then a behavioral study on a reference-free report-writing testbed with a code-generation cross-check, injecting corruption on top of a deterministic reward to isolate the causal channel. Symmetric noise leaves retirement intact, but \emph{false-pass} bias (failures slipping through as passes) disables contribution-based retirement past a sharp threshold (here a false-pass rate of $0.45$) that no amount of data can cross. Separating genuine retirement from cap-eviction churn shows this \emph{mechanism} failure is universal, holding across domains and failure rates and sparing only near-zero-false-pass, verifier-like graders. The downstream \emph{outcome}, though, is regime-dependent: eval quality degrades only where the same corruption also starves skill synthesis, and otherwise holds steady, so the disabled curator is \emph{silent}, surfacing in no aggregate metric. The contribution is a behavioral safety result, not a performance one. A cheap defect-injection audit then tells an operator, before deployment, which side of the threshold their judge occupies.

cs.AI↗
arXiv2026

Side-Chain Tuning of Thermal-Expansion Crossover in Metal-Organic Frameworks

Achieving continuous control over macroscopic thermal expansion remains a fundamental challenge in solid-state physics. Using classical and path-integral molecular dynamics alongside lattice dynamics at near-\emph{ab initio} accuracy, we report an entropy-driven thermal-expansion crossover from positive (PTE) to negative thermal expansion (NTE) in alkoxy-functionalized MOF-5, an archetypal metal-organic framework (MOF). We demonstrate that this non-linear response is continuously tunable via the alkoxy side-chain length, quantified by the number of carbon atoms $n$ grafted onto the archetypal cubic MOF-5 framework: systems with short chains ($n \le 2$) exhibit monotonic NTE, whereas longer chains ($n \ge 3$) trigger a pronounced PTE-to-NTE crossover. At low temperatures, thermal activation of longer side chains opens additional conformational states and generates steric pressure inside the pore, driving positive expansion through a gain in side-chain conformational entropy. Conversely, at elevated temperatures, the side chains enhance transverse linker fluctuations and strengthen the string-tension mechanism associated with low-frequency framework modes, causing structural contraction favored by framework vibrational entropy. Finally, by varying the concentration of side-chain-functionalized linkers, the thermal expansion coefficient can be continuously regulated to realize negative, near-zero, and positive thermal expansion within selected temperature windows. These results establish side-chain engineering as a practical route for programming macroscopic thermodynamic responses in MOFs.

cond-mat.mtrl-sci↗
arXiv2026

Ratchet: How Reliable Must an LLM Judge Be to Retire a Skill?

A large language model (LLM) agent that writes and edits its own skill library must also decide which skills to keep, from one noisy scalar per skill. The answer is exact: a judge scoring failures as passes at rate $(1-\tau)/2$ or above retires nothing, at any sample size, for eviction margin $\tau$. Audits find that machinery is rarely built: LLM-written skills are worth $+0.0$ percentage points (pp) against a no-skill control, human-written ones $+16.2$pp. Unmaintained, a library enters \emph{library drift}, growing until injecting a skill scores worse than injecting nothing. \textbf{Ratchet} repairs this: it evicts each skill on its measured contribution, caps the library at width $C$, and constrains synthesis, lifting held-out $pass@1$ by $+0.328$ on a hard MBPP+ slice. The matching non-divergence bound is finite for exactly two reasons, $C$ and $\tau$. Our contribution is the condition this repair carries and no deployed system states. In reference-free domains the scalar comes from an LLM judge, whose two error directions, modelled as a binary channel, behave nothing alike. Passes scored as failures cost sample efficiency, which more trials buy back; failures scored as passes displace the eviction statistic, and no correction inside the rule recovers it. End-task score is a poor alarm, moving by at most a fifth of the governed lift and not monotonically in the rate. We prove both edges of the certifiable region, confirm them in a running loop, and place a judge on a known side in one offline pass.

cs.AI↗
arXiv2026

Library Drift: Diagnosing and Fixing a Silent Failure Mode in Self-Evolving LLM Skill Libraries

Self-evolving skill libraries face a silent failure mode we term \emph{library drift}: unbounded skill accumulation without outcome-driven lifecycle management causes retrieval degradation, false-positive injections, and performance stagnation. Recent evaluation confirms the symptom (LLM-authored skills deliver +0.0pp gain while human-curated ones deliver +16.2pp (SkillsBench)), yet the underlying mechanism has not been isolated. We provide (1) a reproducible trigger: ablations that isolate drift: one disables skill injection (flat floor, +0.002), one imposes premature retirement (active harm, $-$0.019); (2) trace-level diagnostics: an append-only evidence log with per-skill contribution scores, attribution verdicts, and router engagement metrics that make the failure visible before it reaches end-task scores; and (3) a verified fix: a minimal governance recipe (outcome-driven retirement + bounded active-cap + meta-skill authoring prior) that lifts held-out pass@1 from a 0.258 baseline to a late-window mean of 0.584 (rolling gain $+$0.328) on MBPP+ hard-100 over 100 rounds. Eight ablations decompose which governance mechanisms are load-bearing and which are subsumed, providing a concrete playbook for diagnosing library drift in any self-evolving agent.

cs.AI↗
arXiv2026

MedMIX: Modality-Internal Expert Fusion for Multimodal Medical Diagnosis

Multimodal clinical prediction faces three challenges: multiple foundation models (FMs) with complementary strengths per modality, pervasive missing modalities at training and test time, and sample-specific variation in modality contributions. We introduce MedMIX, a multimodal framework that combines intra-modality expert fusion, learned inter-modality fusion, and training-only large--small model collaboration for robust medical prediction under incomplete modalities. Within each modality, MedMIX aggregates complementary embeddings from multiple small expert models; across modalities, it performs learned fusion over available modalities; and during training, it leverages large teacher models to improve deployed representations without additional inference cost. Across three heterogeneous benchmarks (OpenI, MIMIC-IV-MM, and MMIST-ccRCC), MedMIX achieves consistently strong performance while remaining robust under controlled missing-modality perturbations, and further demonstrates sustained robustness under cross-cohort shift on MIMIC-III. These results highlight MedMIX as a practical framework that unifies within-modality expert collaboration, sample-specific cross-modality fusion, and efficient large--small model collaboration while remaining robust to incomplete modalities.

cs.LG↗
arXiv2026

Hindsight Preference Optimization for Financial Time Series Advisory

Time series models predict numbers; decision-makers need advisory -- directional signals with reasoning, actionable suggestions, and risk management. Training language models for such predictive advisory faces a fundamental challenge: quality depends on outcomes unknown at prediction time. We bridge two ideas from reinforcement learning -- using information unavailable during execution to retrospectively generate training signal, and preference alignment -- and propose Hindsight Preference Optimization: observed outcomes let an LLM judge rank candidate advisories on dimensions that scalar metrics cannot capture, producing preference pairs for DPO without human annotation. We apply this to Vision-Language-Model-based predictive advisories on S&P 500 equity time series, demonstrated by a 4B model outperforming its 235B teacher on both accuracy and advisory quality.

cs.LG↗
arXiv2026

Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agents

As LLM agents scale to long-horizon, multi-session deployments, efficiently managing accumulated experience becomes a critical bottleneck. Agent memory systems and agent skill discovery both address this challenge, extracting reusable knowledge from interaction traces, yet a citation analysis of 1{,}136 references across 22 primary papers reveals a cross-community citation rate below 1\%. We propose the \emph{Experience Compression Spectrum}, a unifying framework that positions memory, skills, and rules as points along a single axis of increasing compression (5--20$\times$ for episodic memory, 50--500$\times$ for procedural skills, 1{,}000$\times$+ for declarative rules), directly reducing context consumption, retrieval latency, and compute overhead. Mapping 20+ systems onto this spectrum reveals that every system operates at a fixed, predetermined compression level: none supports adaptive cross-level compression, a gap we term the \emph{missing diagonal}. We further show that specialization alone is insufficient (both communities independently solve shared sub-problems without exchanging solutions), that evaluation methods are tightly coupled to compression levels, that transferability increases with compression at the cost of specificity, and that knowledge lifecycle management remains largely neglected. We articulate open problems and design principles for scalable, full-spectrum agent learning systems.

cs.AI↗
arXiv2026

Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems

Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku 4.5 (6 methods $\times$ 4 tasks $\times$ 3 repeats), 49% score below zero-shot; on Amazon Nova Lite, the failure rate is even higher. Yet on one task, all six methods improve over zero-shot by up to $+6.8$ points. What distinguishes success from failure? We investigate with 18,000 grid evaluations and 144 optimization runs, testing two assumptions behind end-to-end optimization tools like TextGrad and DSPy, in the order they must be answered: (A) agent prompts interact, requiring joint rather than independent optimization, and (B) individual prompts are worth optimizing at all. Interaction effects are never significant ($p > 0.52$, all $F < 1.0$), and optimization helps only when the task has exploitable output structure: a format the model can produce but does not default to. We further give a mechanistic account: instruction-tuning compresses input phrasing into a narrow output distribution, eliminating the very phrasing-sensitivity that joint optimization assumes. We provide a two-stage diagnostic: an \$80 ANOVA pre-test for agent coupling, and a 10-minute headroom test that predicts whether optimization is worthwhile, turning a coin flip into an informed decision.

cs.AI↗
arXiv2026

Guardrails Beat Guidance: A Large-Scale Study of Rules, Skills, and Persistent Configuration for Coding Agents

Random rules improve a coding agent's task performance as much as expert-curated ones (both $+13.8$pp on a discriminative subset of SWE-bench Verified), and in our data every individually beneficial rule is a negative constraint ("do not refactor unrelated code"), while every individually harmful one is a positive directive ("follow code style"). We arrive at these findings through the first large-scale controlled study of agent rule files (\texttt{CLAUDE.md}, \texttt{.cursorrules}, and the broader family of agent skills, plugin manifests, and persona definitions): we scrape 679 rule files (25{,}532 rules) from GitHub and conduct over 5{,}000 agent runs of Claude Code with Claude Opus 4.6 on SWE-bench Verified. Three patterns emerge. (i) Rule polarity cleanly separates beneficial from harmful rules; we read this through the lens of potential-based reward shaping (PBRS). (ii) Performance gains are largely content-independent: random, shuffled, mismatched-domain, and unconverted-format rule files all match curated rules, pointing to a context priming mechanism. (iii) Individual rules often appear harmful in isolation yet do not visibly accumulate damage in ensemble: pass rates remain stable across rule counts from 0 to 50. These findings expose a hidden reliability risk in the rapidly growing ecosystem of community-authored rules and skills, and they yield a clear principle for safer agent configuration: constrain what agents must not do, rather than prescribing what they should.

cs.AI↗
arXiv2026

The Alignment Floor: How Persona Customization Breaks Safety in Weakly-Aligned LLMs

Telling an LLM to "be enthusiastic" raises its sycophancy rate from 30\% to 50\% on a lightly-aligned model, but has zero effect on a strongly-aligned one. We define this gap as the alignment floor, $\Delta_{\text{floor}}(m)=\max_pS(m,p)-\min_pS(m,p)$, the range of sycophancy rates a model produces across persona conditions, and treat sycophancy as a persona-conditional property rather than a fixed model property. Pluralistic AI relies on behavioral adaptation via persona prompts like "be creative" or "be thorough", which let systems respect diverse user values and communication styles; the safety question is how much customization a given model can absorb before its truthfulness shifts. We present a controlled case study contrasting a strongly-aligned RLHF + Constitutional-AI model (Claude Sonnet 4.6) with a more lightly-aligned model (Amazon Nova Lite), spanning seven persona conditions and five tasks for 1800 total runs. An existence-pair result motivates per-model auditing: there is at least one strongly-aligned model with $\Delta_{\text{floor}}=5$pp (within 5pp of the 15\% control rate) and at least one lightly-aligned model with 45pp (5\%--50\% range). On the lightly-aligned model, all five Big Five personas increase sycophancy over control, and counterintuitively Agreeableness produces the smallest increase, not the largest. The single largest effect in the study is constructive: a Skeptic persona reduces sycophancy by 25pp on the lightly-aligned model, and is the only persona that instructs resistance against user claims rather than engagement with them, suggesting a directionality account. Cross-model transfer of persona effects is near-zero, so persona-alignment testing must be per-model. We propose $\Delta_{\text{floor}}$ as a deployment-time audit metric: measure it on a small persona panel before deploying persona customization.

cs.HC↗
arXiv2026

Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution

We present Verified Multi-Agent Orchestration (VMAO), a framework that coordinates specialized LLM-based agents through a verification-driven iterative loop. Given a complex query, our system decomposes it into a directed acyclic graph (DAG) of sub-questions, executes them through domain-specific agents in parallel, verifies result completeness via LLM-based evaluation, and adaptively replans to address gaps. The key contributions are: (1) dependency-aware parallel execution over a DAG of sub-questions with automatic context propagation, (2) verification-driven adaptive replanning that uses an LLM-based verifier as an orchestration-level coordination signal, and (3) configurable stop conditions that balance answer quality against resource usage. On 25 expert-curated market research queries, VMAO improves answer completeness from 3.1 to 4.2 and source quality from 2.6 to 4.1 (1-5 scale) compared to a single-agent baseline, demonstrating that orchestration-level verification is an effective mechanism for multi-agent quality assurance.

cs.AI↗
arXiv2026

AdaField: Generalizable Surface Pressure Modeling with Physics-Informed Pre-training and Flow-Conditioned Adaptation

The surface pressure field of transportation systems, including cars, trains, and aircraft, is critical for aerodynamic analysis and design. In recent years, deep neural networks have emerged as promising and efficient methods for modeling surface pressure field, being alternatives to computationally expensive CFD simulations. Currently, large-scale public datasets are available for domains such as automotive aerodynamics. However, in many specialized areas, such as high-speed trains, data scarcity remains a fundamental challenge in aerodynamic modeling, severely limiting the effectiveness of standard neural network approaches. To address this limitation, we propose the Adaptive Field Learning Framework (AdaField), which pre-trains the model on public large-scale datasets to improve generalization in sub-domains with limited data. AdaField comprises two key components. First, we design the Semantic Aggregation Point Transformer (SAPT) as a high-performance backbone that efficiently handles large-scale point clouds for surface pressure prediction. Second, regarding the substantial differences in flow conditions and geometric scales across different aerodynamic subdomains, we propose Flow-Conditioned Adapter (FCA) and Physics-Informed Data Augmentation (PIDA). FCA enables the model to flexibly adapt to different flow conditions with a small set of trainable parameters, while PIDA expands the training data distribution to better cover variations in object scale and velocity. Our experiments show that AdaField achieves SOTA performance on the DrivAerNet++ dataset and can be effectively transferred to train and aircraft scenarios with minimal fine-tuning. These results highlight AdaField's potential as a generalizable and transferable solution for surface pressure field modeling, supporting efficient aerodynamic design across a wide range of transportation systems.

cs.CE↗