Papers

Agent benchmarks usually measure task completion and treat resource use as an auxiliary statistic. In deployment, however, the choice among a local lookup, broad search, composite research tool, stronger model, or human escalation is part of the task itself. We introduce EcoAgent-Bench, in which every task specifies priced actions and an explicit budget. Its 304 real-derived tasks span five families adapted from GAIA, HotpotQA, and MuSiQue, and test four decisions: avoiding unnecessary escalation, escalating when local evidence is insufficient, selecting a model tier, and stopping on unsupported premises. We evaluate seven LLM agents in tool-API and workspace-CLI settings, together with four oracle scripted controls. Micro-averaged accuracy rewards one-sided policies: always-escalate controls achieve high micro success while failing save-oriented tasks. We therefore also report an economic-consistency score (the worse of accuracy on upgrade-oriented and save-oriented family groups) which exposes this failure. Tool-API agents attain only 3.9-24.0% micro strict success (at most 7.3% economic consistency), often either stopping before warranted escalation or overspending on cheap tasks. A threshold-crossing budget sweep changes GPT-5.4's escalation rate from 0% to only 3%. These results show that completion under a budget and economical action selection are distinct properties. We release the task bundle, transformation pipeline, frozen evaluation environments, and integrity-bound result artifacts needed to study both.

Modern deep neural networks achieve strong performance, but their scale makes them costly and slow, especially on resource-constrained edge devices. Pruning and quantization address this, but rely on manual, expert choices and on algorithms that are hard to apply across architectures. Uniform settings also ignore how differently individual layers respond to compression, which costs accuracy. We introduce APQF, an agentic profiling-guided framework that combines structured pruning, mixed-precision quantization-aware training, and accuracy recovery in one automated pipeline. A profiling agent measures how cost is distributed across the model and how sensitive each part is to pruning, and this evidence drives per-layer pruning ratios, per-layer bit-widths, and the recovery strategy, all proposed by LLM planners and validated before execution. To our knowledge, APQF is the first framework to combine LLM-guided, profiling-grounded decisions with a fully training-aware pruning and quantization pipeline for both CNNs and vision transformers. We evaluate APQF on ResNet, VGG7, ViT, DeiT, and Swin using ImageNet-1k and CIFAR-10. On ImageNet it cuts compute to 5.6-7.7 percent of the original bit-operations, a 13-18x reduction, while keeping accuracy close to the baseline, and under a 200K-image budget it stays roughly 17 points higher in Top-1 than existing joint pruning and quantization methods. On CIFAR-10 it compresses further than that method on four of five architectures. On VGG7 it reaches 93.15 percent using only 0.41 percent of baseline bit-operations, the only method at that compression level to improve on its full-precision baseline. Ablations show that uniform compression loses the most accuracy at matched compute, and that withholding profiling data from the planner hurts every model. Six LLM planners, including free open-weight ones, all reach 97.4-97.9 percent on Swin-Tiny.

Language models (LMs) are increasingly used to interact with external services via programs written in domain-specific languages (DSLs). Unfortunately, since DSLs are often low-resource and esoteric, LMs frequently produce syntactically invalid programs in these languages. Grammar-constrained decoding can eliminate such failures, but requires syntactic constraints. These are usually in the form of a context-free grammar for the target language, an artifact that is hard to come by for third-party DSLs. In this work, we define an agent, called Autogrammar, that automatically learns context-free grammars from documentation and execution data. Autogrammar is formalized as a Kripke structure whose nondeterministic choices are resolved by a language model, enabling declarative control of agent behavior via linear temporal logic constraints. We evaluate four versions of Autogrammar on three DSLs (i.e., Amazon CloudWatch Logs Insights, Dynatrace Query Language, and Datadog Search Syntax) and find that it generates grammars that achieve near perfect precision on unseen data; that temporal restrictions reduce execution time by 3.8x without incurring statistically-significant loss in precision; that execution data is crucial while documentation is dispensable; and that grammar-constrained decoding using Autogrammar-generated grammars significantly improves end-to-end LM performance on eight out of ten real tasks, matching or exceeding the performance of a professionally-maintained grammar. In comparison, the context-free grammars generated by existing LM baselines and a state-of-the-art formal technique perform significantly worse over the same evaluation.

Autonomous agents now carry out entire data analyses, selecting cohorts, joining tables, and fitting models with little step-by-step supervision. When such an analysis turns out to be wrong, someone must determine which operation caused it. A recent approach does this without any labelled mistakes, learning instead from analyses known to be sound and flagging operations that depart from what that model predicts; how reliable such audits are has not been studied. This paper supplies that analysis. The choice of score determines whether an error can be localized at all. If each operation is scored by how surprising it is given the operation immediately preceding it, then operations that merely inherit an earlier error are indistinguishable from correct ones, so one mistake produces one flag; scores computed against a longer reconstruction of the intended analysis instead spread a single mistake across many operations. We quantify how far they spread, and how to choose the comparison length when an error accumulates gradually rather than at once. We then give procedures that control the proportion of falsely flagged operations within a single audited analysis, requiring only that sound analyses be exchangeable rather than that the fitted model be correct, and we quantify how much the guarantees weaken when the model is imperfect or when the analysis was selected for review in a way that depends on its content. Finally we establish a limit on what any such audit can report: errors below a certain magnitude cannot be attributed at all, being indistinguishable from ordinary variation among sound analyses. This limit falls so slowly as more sound analyses are collected that at the representation sizes now in use a hundredfold increase reduces it by under two percent, so the dimension of the representation rather than the volume of training data is the binding constraint.

Large language model agents can discover alphas, yet current methods have three weaknesses. The search cannot adapt during the run, automation usually ends at alpha generation while library selection and model choice stay manual, and alpha discovery can read the test window through loop feedback or code problems. We present AutoScientist-Quant, a self evolving search process that regards quantitative research as one budgeted search problem. A single controller conditions every decision on the remaining budget, choosing at each round whether to improve, combine, pivot, or stop, which node to expand, how many alphas to generate, and how to retrieve past trajectories from the shared memory. The same core then selects from the library and tunes the model, closing the loop from hypothesis to deployable strategy. We also review the evaluation pipeline reused from prior work, fix two lookahead problems, and keep the feedback window disjoint from the held out test window, so every comparison tests true generalization. On CSI universes, the framework attains the best value of nearly every metric in every setting, and these conclusions hold across several backbones and markets.

An AI scientist should not grade its own homework. Yet in the systems we examined, the agent that reviews the work usually comes from the same model family as the agent that produced it, or at least from the same vendor. Model evaluators are known to favour their own generations. Whether models trained alike also share blind spots is a conjecture, not a settled finding, but if they do, the reviewer inherits the author's. The record of what was flagged and what was waved through often sits in platform logs that nobody outside can replay. We present CrossAudit, a protocol for supervising autonomous research pipelines. It rests on three commitments. Each increment of work is audited by an agent from a different vendor against a rulebook a human wrote and versioned. Reports, verdicts, disputes and rulings are git commits, so the supervision history can be re-read and cited; raw model exchanges are not yet part of that record. Scripted checks run before any model does. Advisory judgement never gates the pipeline: a model blocks only by citing a rule, and no model may waive a deterministic failure. Blockers that survive a bounded number of revision rounds go to a person. We state the protocol as eight invariants. We describe a reference implementation built from GitHub Actions and a few hundred lines of Python, and report a live deployment of a closely related variant in a computational-chemistry pipeline. We also ran a seeded-defect trial (30 increments, 43 seeded defects, one run per configuration). A cross-vendor audit of our own repository then voided its blinding. We adopt that audit's findings and report the corrected results. The trial shows that two vendors read the same rulebook differently. It does not show that either is better. The strongest evidence here is the committed, uncontrolled record of cross-vendor audits of this paper itself.

A trained reinforcement learning policy does not determine the complete behavior that users encounter: deployment code still schedules, admits, suppresses, or replaces its proposed actions. We contribute runtime action interference (RAI), an AI control mechanism that preserves policy parameters while regulating action pacing and filtering configured action patterns after inference. RAI releases a proposed action only when its cooldown condition is satisfied and its content detector does not flag the action; otherwise, it dispatches a no-op. The detector covers specified toxic behaviors, including worker-unit harassment, while the cooldown controls action rate. We implement RAI in a replication of AlphaStar actor.py and make the implementation and reproducibility materials available through an open source code repository. We deployed RAI in a StarCraft~II human participant study that compared two presentations of the same opponent with high capability and rate limited actions; we withheld its capability claim in one presentation and disclosed it in the other. On response scales from 1 to 5, we observed pooled fairness, trust, and toxicity means of 3.90, 3.50, and 2.00 under claim withholding, compared with 2.62, 4.31, and 2.85 under disclosure. Disclosure corresponded with lower perceived fairness and higher perceived toxicity across every expertise group, whereas trust increased among novices and experts but decreased among intermediate participants. Our human evaluation therefore shows that perceptions of an opponent controlled through RAI can vary substantially with the capability information presented to users, even when the configured control remains constant. We conclude that human-computer evaluations must separate control within the execution stack from capability disclosure and assess fairness, trust, and toxicity as distinct dimensions of human experience.

Existing multi-task learning methods rely on hard sharing, multiple paths or experts, adaptive sharing, and dynamic expansion. However, their capacity changes are usually constrained by predefined structures or triggered by task boundaries and conflict signals. This raises a fundamental question: can a network start from exact single-path computation and grow a new independent path only when persistent optimization evidence appears? We propose the Emergent Modular Atomic Network (EMAN), an optimization-driven framework for exposing an antisymmetric growth direction through latent relative phases without instantiating a second path, and for monitoring multiple decision signals during training to transform local optimization evidence into a structural decision. EMAN materializes two equal-capacity independent paths only after certification. EMAN adaptively allocates shared and task-specific representation capacity to accommodate varying task requirements. Extensive experiments on controlled rank settings, PASCAL-Context, and NYUv2 validate its effectiveness, achieving improved performance at a competitive computational cost.

LLM judges increasingly evaluate responses against fine-grained rubric checklists. When a sample requires multiple rubrics, current methods typically assess each in a separate inference call. Evaluating all rubrics in a single pass is a natural alternative with greater efficiency, but we find that it introduces rubric interference: the verdict on one rubric shifts depending on which other rubrics are co-present. In a preliminary study, only one-third of samples receive fully consistent verdicts when evaluated under rubric sets of varying composition. We develop a measurement framework that probes interference through four controlled operations: rubric set expansion, subsetting, reordering, and noise injection. To mitigate interference without external supervision, we propose Self-Anchored Rubric Alignment (SARA). SARA uses a model's own single-rubric judgments as stable anchors and aligns multi-rubric reasoning with these anchors through on-policy self-distillation. We validate SARA on three datasets (HealthBench, FLASK, ResearchQA) and two model families (Qwen3, Llama-3.1). SARA consistently improves evaluation consistency while maintaining agreement with both base models and GPT-4.1 as a reference judge. Furthermore, the learned consistency transfers across datasets, confirming that SARA teaches a general capability rather than fitting dataset-specific patterns.

Words recur constantly in natural language use, yet it remains unclear whether language models reactivate prior representations or re-evaluate repeated words afresh, and whether post-training changes this default behavior. We apply repetition priming (Shiffrin and Schneider, 1977) to 15 models across five model families (1.5B-14B parameters) in two tasks, semantic categorization and cloze completion, with matched human experiments using identical stimuli. We find that base models exhibit automatic processing: they show immediate facilitation that remains stable across lags, partially survives context removal, and correlates with attention to prior occurrences. Instruct models exhibit controlled processing: their facilitation decays with lag, collapses without expected context, and reverses to interference at larger scales. Within the Qwen 2.5 family, this dissociation increases monotonically with model scale, suggesting that post-training progressively alters repetition processing. Humans show a hybrid profile, with lag-sensitive facilitation resembling instruct models but without interference, suggesting that neither model type fully captures human cognition. Our findings reveal a qualitative shift in how language models process repeated information after post-training and provide mechanistic evidence for the divergence between model behaviors.

Context-aware recommender systems have long recognized that factors such as location, time, and weather shape where and what people choose to eat. Existing weather-aware food and point-of-interest recommenders, however, typically treat weather generically – mapping conditions to preferences through hand-crafted rules or specially trained context models – and do not capture that the culturally appropriate response to weather is itself region-specific: a rainy evening calls for hot tea and fried snacks in one culinary culture and for very different comfort food in another. Encoding such weather-by-region-by-cuisine interactions as explicit rules or training data is brittle and does not scale. We present a weather- and location-aware agentic dining-recommendation system that takes a different approach: a large language model (LLM) orchestrates tools for location and weather retrieval and then reasons in natural language over the combined context, drawing on the cultural and culinary world knowledge already latent in the model to produce region-sensitive, weather-appropriate recommendations without per-region rule tables or specialized training. We describe the agent architecture, the tool-orchestration flow (Google location services and a weather service feeding an OpenAI LLM), and the reasoning mechanism, and we report on a working prototype that was implemented and briefly deployed end-to-end. We discuss design trade-offs – cost, latency, ambiguity handling, and fallbacks – and we are explicit about limitations, including the absence of a formal user study and the risk of cultural stereotyping in locality-based inference. The contribution is architectural: a simple, extensible pattern for incorporating environmental and cultural context into agentic recommendation through LLM reasoning rather than engineered rules.

Group-based reinforcement learning methods for multimodal large language models typically rely on trajectory-level credit assignment that applies a single advantage to all tokens in a response. However, multimodal reasoning involves substantially higher perceptual uncertainty than text-only settings, where the model must repeatedly re-examine visual information to verify intermediate interpretations, and different visual groundings can lead to divergent reasoning paths, making such uniform credit assignment particularly inadequate and causing relative advantages to progressively degenerate toward zero. To address these challenges, we propose Multi-Branch Policy Optimization (MBPO), a tree-based framework that constructs reasoning trees at vision-language decision boundaries, enabling sibling branches to explore diverse visual hypotheses and assigning segment-level credit through branch-relative advantages. We further introduce a temporal replay buffer to reuse informative segments while controlling policy staleness. Experiments on several multimodal reasoning benchmarks show that MBPO outperforms representative baselines, improving both learning signal quality and optimization efficiency. The code is publicly available at https://github.com/ShuaiLyu0110/MBPO.

Perception in biological systems is inherently noisy, requiring organisms to make decisions under uncertainty where misclassification can be costly or fatal. We present an Artificial Life predator-prey model of foraging under noisy perception, and compare agent performance when using various policies that take into account their noisy predictions. Through controlled experiments under both symmetric and asymmetric perceptual noise, we show that blindly trusting perceptual labels leads to catastrophic failure as noise increases, while uncertainty-aware strategies significantly improve survival and reduce fatal errors. We further observe qualitative regime shifts in behaviour, with agents transitioning from exploratory to conservative strategies as uncertainty increases. Our model links risk-sensitive foraging, ecological information use, and Artificial Life by showing that explicit information gathering can improve robustness when perception is unreliable. These results highlight the importance of uncertainty-aware decision-making and provide an interpretable artificial life analogue to robust learning with noisy labels.

Watermarking traces the provenance of text produced by large language models by embedding statistically detectable signals during decoding. Existing schemes fall into logits-based, sampling-based, entropy-aware, and adaptive-strength families, yet all of them place watermark signals according to local token statistics. In the open-ended text-generation settings evaluated in this work, local statistics may provide insufficient guidance for placing robust watermark signals. We introduce WorldMark, a plug-and-play interface that uses World Knowledge Memory (WKM) to organize semantic and episodic knowledge in a memory graph, converts the retrieved knowledge into a token-level knowledge saliency score, and adjusts the strength of a host watermark through Asymmetric Knowledge Modulation (AKM). WorldMark requires no backbone retraining and introduces no additional detector-side model or parameter. On the primary C4 evaluation, the complete WorldMark interface improves clean and attacked detection across three adaptive-strength host variants while slightly reducing perplexity. Additional pilot experiments on C4 and OpenGen show that direct memory conditioning transfers across multiple watermark families but can be unstable without saliency-aware modulation. WorldMark requires no additional detector-side model or parameter and introduces negligible overhead under the primary protocol.

Multi-head attention combines an input-dependent softmax routing with an input-independent linear value projection, so the per-sample operator mapping aggregated values to outputs is the same for every input set. We study the consequences of this asymmetry for permutation-invariant set targets. We introduce the Transformation Degrees of Freedom (TDOF) of a target operator, a complexity measure counting the input-dependent directions an exact representation requires, and present a depth-separation analysis showing that context-rigid attention needs depth proportional to the target's TDOF, whereas a single layer with a context-adaptive value family can represent the same target. Building on this analysis, we propose Matrix Zonotopic Attention (MZAttn), which replaces the fixed value projection with a context-adaptive matrix-zonotope family: a centre matrix plus a sum of generator matrices weighted by input-dependent gates. The construction reduces to standard multi-head attention at initialisation, preserves permutation equivariance, and admits a data-driven reachability interpretation. Experiments on a range of set-prediction tasks are consistent with the TDOF prediction that the architectural advantage is selective: it appears on targets that depend on the input set in a high-rank, sparsely combinatorial way, and is small on aggregate-statistic targets where parameter-matched standard attention is already competitive.

High-quality long-horizon training data for terminal agents is expensive to produce, often costing hundreds to thousands of dollars per task, because each task must keep the instruction, environment, reference solution, and verifier mutually consistent. Human authoring does not scale, and direct generation with large language models (LLMs) often breaks these dependencies. We present Recursive Synthetic Terminal Tasks (RST), a recursive verified synthesis framework for constructing long-horizon terminal-agent tasks at scale. Starting from verified seed tasks, RST extends the reference solution, realigns the verifier and instruction to the new workflow, validates the result in a fresh sandbox, and reuses accepted tasks as seeds for subsequent rounds. Across fifteen recursive rounds, RST produces 37,484 synthesized terminal-agent tasks at roughly $0.05 per task. Task difficulty increases substantially over rounds: the median reference solution grows from 67 to 374 lines, the median number of executed commands grows from 40 to 244, and DeepSeek-V4-Pro pass@4 drops from 90% at R1 to 2.5% at R15. To demonstrate training utility, we collect rejection-sampled Qwen3.5 trajectories on the synthesized tasks and use them for supervised fine-tuning. Fine-tuning on these trajectories improves Qwen3.5-27B and Qwen3.5-122B-A10B by up to 10 points on Terminal-Bench 2, Terminal-Bench Hard, and Long-Horizon Terminal Bench, while agentic PPO lifts Qwen3.5-27B to 49.44%, 32.00%, and 22.07% on the three benchmarks, corresponding to relative gains of 20.0%, 41.2%, and 21.9% over the base model. Moreover, after 15 rounds, the recursion shows no ceiling: synthesis yield and validation rates remain stable as difficulty keeps climbing, indicating that the process can continue well beyond the scale reported here.

When a student must learn concepts connected by prerequisite dependencies, when does the order of instruction matter, and what does it cost to find the best one? We study instructional sequencing as a stochastic shortest-path problem in which attempting a concept succeeds with a state-dependent probability and failure leaves the learner state unchanged. We first prove that this stochasticity can be eliminated exactly: the problem collapses to a deterministic shortest-path problem on the lattice of prerequisite order ideals, preserving optimal values and actions. The collapse removes stochastic complexity but not combinatorial complexity: optimal sequencing remains NP-hard – via reduction from feedback arc set in tournaments – even with no prerequisite edges, unit costs, uniform binary nonnegative transfer, and success probabilities at least $1/2$. Hardness is not uniform: when realizable transfer preferences remain jointly acyclic with the prerequisites, any topological order of the residual joint graph is optimal, and fixed prerequisite width yields polynomial-time exact dynamic programming. A computable diagnostic, $m\Delta$, bounds the value of sequencing before optimization. On 70,893 interactions from an introductory CS course, the diagnostic certifies a doubly easy regime – little value to optimize and little space to search – while constructed transfer instances realize the challenging regime, where myopic sequencing suffers large regret yet exact A* with a consistent heuristic expands only linearly many states on that family.

Translating natural language instructions into machine-interpretable formal specifications enables robots and autonomous systems to plan, reason, and formally verify their behavior. However, existing translation models typically generate a specification for every input, even when the result is unreliable or fails to capture the user's intent, creating risks in safety-critical applications. Inspired by selective conformal prediction, we propose a selective translation framework that not only generates formal specifications but also determines when they can be trusted. Reliability is scored by two complementary black-box signals, the fidelity of the specification back-translated into natural language and the dispersion of repeated translations under exact semantic equivalence, which fail on different errors and jointly separate incorrect translations more sharply than either alone. Conformal risk control calibrates this score into a decision that accepts a specification or abstains, with a distribution-free bound on the rate at which incorrect specifications are accepted for execution, and a conformal anomaly detector on instruction embeddings screens out-of-distribution inputs before any translation is attempted. The proposed framework is general across formal specification languages, with experiments on Signal Temporal Logic (STL), Linear Temporal Logic (LTL), and geometric Spatio-Temporal Logic (SpaTiaL) demonstrating improved translation reliability, robustness under the evaluated cross-tier shifts, and effective uncertainty-aware abstention. This work establishes a foundation for trustworthy natural language interfaces by enabling AI systems to recognize when generated specifications may not be reliable.

The life sciences and health research have started to benefit from artificial intelligence, which raises ethical concerns that are real but, we argue, not special. Any science should be governed by values that rest on how the human brain is built and socialised rather than anything distinct to artificial intelligence. Importantly, the human brain has a different, much less costly computational architecture than these machines. This is achieved through the orchestration of a global neuronal workspace, and through reward best described not as a quantity to be maximised but as a continuous cycle of wanting, liking and satiety. As such, this creates the deep tension running through the ethics of the human person, between the universality of ethical judgement and the diversity of morals. The brain networks of the global workspace and emotion are universally shared, but the diversity of content is shaped by epigenetic appropriation of the particulars of the physical, social and cultural world, which makes every person unique. Still, if we were to build machines on these principles rather than the present unaffordable reward maximisers, the question of their governance would change from restraint to upbringing. We set out the institutions such a future would require, together with the questions that remain open.

In generative training, a model produces an output and is penalised for its difference from an example. With one output per comparison, a model that produces one common answer can outperform a model retaining a broader repertoire. Explorative Modeling (XM) produces $K$ outputs per comparison and updates on the closest, claiming exploration as a "third pretraining axis" associated with generative expressivity. Here I show the third axis is actually freedom, meaning the weakness of the constraint implied by a model's behaviour. Previous work showed freedom is a property of function rather than form. Parameters, architecture, minimum-description-length (MDL), and data can vary while the behavioural constraint remains unchanged. It was formally proved that weakest models are likeliest to generalise, and freedom selection beat MDL by 110-500% in induction experiments. I prove average XM loss depends on the chance a candidate misses an acceptable region, with exploration raising miss probability to power $K$. For $K>1$, match probability rises with freedom. I then demonstrate empirically that XM optimises for freedom. In a Forward XM experiment, larger $K$ increased or saturated measured freedom, and increased freedom at every tested value under context-dependent targets. I trained XM candidate pools and compared validation selection with a freedom selector that read unlabelled parent contexts. Freedom won in 29 of 30 cases. Generative expressivity is a mode-count proxy for freedom, that discards the extension structure that gives freedom its generalisation significance. XM is a means, freedom an end, and selecting for freedom improved XM under distribution shift.

Models trained by empirical risk minimization on data containing spurious correlations achieve high average accuracy while failing on subpopulations where the correlation does not hold. Existing methods for identifying the affected samples without group annotations rely on signals from early training, which requires locating the epoch at which to intervene, a hyperparameter typically selected using group-labeled validation data. We show that a usable signal is available after convergence, when loss no longer distinguishes the two populations. Samples consistent with the spurious correlation are classified by a shared rule, while the remaining samples are fit through configurations specific to individual inputs and are correspondingly more fragile. Applying a fixed perturbation to a converged model's inputs flips the predictions of the latter far more often than the former. The resulting procedure requires two forward passes per training sample, no group annotations at any stage, and no early-stopping epoch. Using the detected samples to rebalance training raises worst-group accuracy on Waterbirds from 57.3% to 80.8%, against 85.8% with ground-truth group labels.

AI tools are being increasingly adopted in policing in the UK and worldwide. Racial bias is a known and well-documented risk, yet representatives of affected communities are rarely included in decisions about AI adoption. We present results from a mixed-stakeholder deliberation workshop bringing together 30 community representatives, police officers, and academics to assess the risks of 13 AI use cases in policing, with an explicit focus on racial bias. We found that participants were broadly open to AI adoption, rejecting only three use cases outright – most notably recidivism risk assessment, where objections targeted the premise rather than the implementation. Our analysis reveals that foregrounding racial equity did not narrow the deliberation. Instead, discussions gravitated toward a fundamental set of questions: does this tool actually work, will it deliver genuine benefit, and will that benefit extend to everyone? This integrated reasoning—reminiscent of the curb-cut effect in inclusive design—highlights the benefit of incorporating the racial bias lens into the risk-benefit analysis of AI use cases from the outset.

Large language models (LLMs) are increasingly used as AI tutors, but a correct answer is not always a pedagogically appropriate one. In classroom learning, effective help depends not only on correctness, but also on whether a response matches the learner's current foundation, the course sequence, and the timing of concept introduction. Existing evaluations focus mainly on answer quality, leaving this instructional fit under-measured. We present the Pedagogical Suitability Index (PSI), a composite metric of six theory-informed sub-scores that evaluates how well LLM-generated tutoring responses align with learner readiness and curricular progression, and we further use PSI as a structured feedback signal for response improvement. We evaluate four LLM tutors (ChatGPT, Gemini, Gemma4, and Qwen3) across 240 scenario-based evaluations using paired standard and defective prompts, then apply a PSI-guided regeneration protocol to 62 weak-performing cases. Baseline differences across the four tested models were modest overall (PSI range: 0.557 to 0.638), and open-weight and closed models did not exhibit a clear separation in pedagogical fit. Under the tested prompt perturbations, overall PSI remained largely stable (Delta = -0.002), though sub-score trade-offs emerged. More importantly, PSI-guided feedback substantially improved weak-performing cases: 51 of 62 cases improved (82.3%). Focused manual evaluation of the 62 PSI-selected weak cases provides initial evidence that the identified weaknesses are instructionally meaningful and that many PSI-guided regenerations correspond to human-judged improvement. These results suggest that learner- and curriculum-aware alignment may matter more for effective tutoring than model category alone, and that such alignment is both measurable and improvable.

Care plan coordination demands synthesizing heterogeneous clinical, functional, and psychosocial information across multiple professional disciplines, where monolithic LLM pipelines cannot perform in a transparent or safe manner. We present CANOE (Contestable Argumentative Network-of-Experts), a multi-agent neuro-symbolic framework that addresses these limitations through five modules: complexity assessment, adaptive team recruitment, role-based argumentative computation via an Arena-based Quantitative Bipolar Argumentation Framework (A-QBAF), human-in-the-loop contestation, and care-plan synthesis. Role-specialized agents generate supporting and attacking arguments for candidate interventions; conflicts are resolved through arena-based clash resolution before acceptability scores propagate across the argumentation graph. Care planners may accept, reject, edit, or add arguments, and the framework will deterministically recompute the final plan. Evaluation on Discharge Me! and MedicalRAG using ROUGE-L, AlignScore, MEDCON F1, FKGL, and LLM-as-a-judge shows that medically fine-tuned models achieve the strongest clinical correctness and safety, while CANOE's argumentative structure provides faithful explanation and human contestability.

Language models can prepare VASP INCAR files from natural-language requests, but so far only large proprietary cloud models come close to handling the tightly coupled, physics-sensitive settings reliably, a dependence that fits poorly with local, high-throughput materials workflows where privacy, cost, and offline deployment matter. We show that a small language model (SLM) can close this gap. The SLM is fine-tuned on reference VASP calculations and paired with VASPGuard, a deterministic post-processor that checks syntax, workflow, and material-dependent constraints; we call the combined model INCAR-SLM. On INCARBench, a benchmark for VASP INCAR generation, INCAR-SLM built on Qwen3-4B outperforms every general-purpose LLM evaluated, exceeding GPT-5.4 by 15.55 points on the 100-point INCAR Score. Most of this gain comes from fine-tuning, with VASPGuard correcting the errors that remain. We further find that model size matters less than expected: once fine-tuning and post-processing are applied, performance saturates at a few billion parameters, and Qwen3-4B outperforms larger models in the same family.

Current Multimodal Large Language Models (MLLMs) can process diverse sensory inputs, yet their reasoning remains heavily biased toward a dominant modality, resulting in brittle cross-modal reasoning. We introduce C$^3$PO, a benchmark of 3,404 samples spanning video, audio, image, and text, evaluating two abilities: information composition (fusing dispersed evidence) and counterfactual conflict (resolving deliberate contradictions). C$^3$PO's paired IC/CC structure and four-tier design enable targeted diagnosis of when and why cross-modal reasoning fails. Built through a fully automatic pipeline using 25 logically grounded templates, C$^3$PO reveals that while humans achieve 88.64% accuracy, the best model (Gemini-3.1-Pro) reaches only 73.17%, with open-source models collapsing under conflict. Through attention probes, we find 86-95% of failures stem from modality dominance: models commit to one modality while ignoring contradictory evidence, concentrating 87-95% of attention on text. Mid-layer attention entropy predicts correctness-sustained exploration succeeds, premature collapse fails. The 56-point accuracy gap between equally complex templates reveals that performance depends on modalities' structural roles in conflict resolution, not combinations. These findings show multimodal perception does not guarantee robust reasoning; architectures must enable sustained cross-modal attention to avoid premature

Clinical machine learning (ML) has the potential to support high-stakes medical decision-making, but reliable deployment is often constrained by scarce, heterogeneous, and temporal complexity. Developing effective ML pipelines for such data remains time-consuming and error-prone, while existing automated machine learning (AutoML) systems only partially address this challenge because they largely rely on brute-force search over predefined spaces and lack explicit reasoning and memory. We therefore reformulate AutoML for small clinical data from exhaustive search to reasoning-driven refinement. We propose DoctorAgents, an agentic AI framework that autonomously constructs and optimizes end-to-end ML pipelines through specialized large language model (LLM) agents for generation, validation, and refinement. DoctorAgents backpropagates natural-language feedback through textual gradient descent to perform targeted updates without exhaustive search. Experiments across diverse clinical tasks show that DoctorAgents consistently outperforms established AutoML baselines while producing more interpretable task-specific representations.

Counterfactual analysis aims to predict potential outcomes under hypothetical scenarios, offering valuable insights for decision-making. This paper investigates the application of large language models (LLMs), specifically the GPT-3.5 model, for counterfactual analysis. We focus on the online lending context, where the counterfactual return on investment (ROI) is crucial for evaluating different interest rate schemes. We begin by assessing the predictive performance of GPT and comparing it with advanced machine learning algorithms. The results show that prompt engineering can significantly enhance GPT's predictions, with the R-squared increasing from 1.97% to 2.84%, closely approaching the 3.48% achieved by gradient-boosted regression. Subsequently, we utilize GPT to generate counterfactual ROIs under a set of alternative interest rates. GPT exhibits logical coherence and causal reasoning in its responses. The findings underscore the potential of LLMs as effective tools for counterfactual analysis in online lending, suggesting broader applications for LLMs in various predictive and decision-making contexts.

CASCADE is an agentic framework that predicts downstream transcriptional effects of gene perturbation from precomputed ARACNe regulatory networks, exposed via MCP. Prior work validates such tools by checking whether predicted genes are known cancer genes (membership); we instead test whether the predicted direction of change matches reality, using focal-gene copy-number amplification as a dosage-based proxy for the inverse of knockdown against real TCGA patient tumor data. For MYC, CASCADE's predicted knockdown targets show strong concordance with real amplified-vs-non-amplified tumor expression across three cancer types (BRCA: 90.0%, COAD: 72.0%, STAD: 85.7%; all p<=0.0013), well above permutation baselines, surviving a PAM50 subtype control and replicating in an independent cohort (METABRIC, 87.2%). Compared against curated MSigDB gene-set baselines via Fisher's exact test, CASCADE's accuracy is not shown to exceed existing public knowledge of MYC- or E2F-driven biology, though its gene-specific direction-calling clearly outperforms a naive uniform guess. Extending to fifteen additional genes, validation proves gene-specific rather than universal: proliferation-machinery regulators mostly replicate, while lineage-identity transcription factors and one cyclin-D paralog (CCND2) consistently fail, a pattern we discuss as a hedged, post-hoc hypothesis. We separately benchmark whether an LLM-based agent correctly grounds natural-language requests into CASCADE's real MCP tool calls. Across 35 queries, a documented local model reaches 71.4% exact match (85.7% for a larger model); schema and gene-alias failures are resolved by scale or server-side correction, but both models confidently default to the wrong perturbation type on ambiguous queries, a failure a targeted fix could not resolve because its trigger condition never occurs.

Time-sensitive networking (TSN) is increasingly integrated into mobile edge computing (MEC) to support applications with stringent latency requirements, such as extended reality (XR). However, existing TSN scheduling solutions predominantly rely on static optimization techniques or centralized learning models that are based on fixed traffic patterns, limiting their effectiveness in dynamic environments. In practice, MEC environments often host multiple co-located XR traffic flows whose characteristics evolve over time, creating complex inter-queue dependencies that current schedulers fail to capture. Addressing these challenges requires adaptive, decentralized scheduling mechanisms capable of coordinating multiple TSN queues under varying traffic conditions. To this end, this paper proposes a multi-agent reinforcement learning (MARL) framework for TSN scheduling, where each TSN queue is modeled as an autonomous agent. The Heterogeneous-Agent Proximal Policy Optimization (HAPPO) algorithm is employed to explicitly model inter-agent dependencies and jointly optimize service delivery across queues. The simulation results demonstrate that the proposed approach reduces average frame waiting times by up to 26.8% and worst-case delays by approximately 16.8%, highlighting its effectiveness in dynamic XR-driven MEC scenarios.