Skip to main content
Search

Purdue University West Lafayette

17 related articles

Subscribe to updates
  1. arXiv

    ARoute decides which model answers before any call: highest mean accuracy in all 15 task-cost settings, 4.1 points over the strongest baseline at matched cost

    ARoute predicts both whether upgrading to the stronger model will help and how much more that upgrade will cost, using the query text alone, and picks the model before either model is invoked. It fits two lightweight heads on a small set of paired records over one frozen BERT encoder. On MMLU-Pro, LiveCodeBench, SuperGPQA and two SWAN value-filling tasks, ARoute-Batch attains the highest mean accuracy in all 15 task-cost settings, 4.1 points above the strongest baseline on average and up to 11.9 points, and completes a batch up to 2.8 times faster than cascades.

  2. bioRxiv

    Genes no longer queue by index in attention: SPEAR folds co-expression geometry and perturbation-time repositioning into rotation angles

    Transformer models for single-cell perturbation prediction need a position for each gene, yet genes have no natural order, and static co-expression masks cannot reflect the regulatory rewiring a perturbation induces. SPEAR replaces that position with continuous coordinates from the co-expression graph Laplacian and, at every layer, re-derives a scalar position from the perturbation-conditioned hidden state, summing both into one rotation angle. On genetic, pharmacological and cytokine perturbation benchmarks it beats baselines including scDFM on differential-expression direction and magnitude, with the largest gap on held-out combinations.

  3. arXiv

    A glance is enough to queue the next actions: symbolic graphs carry kitchen priors, and a 3M-parameter model leads in short-observation tests

    SymAnt merges a scene graph and a knowledge graph into one joint graph and uses the symbolic context found by graph search as the starting point for iterative decoding, so it can predict long action sequences from very short observation windows. In the short-observation settings of Breakfast and 50 Salads, its mean accuracy over classes exceeds the neuro-symbolic baseline NeSCA and several large-model baselines; on EPIC-KITCHENS, the 3M-parameter model performs comparably to fine-tuned models with billions of parameters. Because predictions are anchored in an editable graph, users can add or remove nodes at inference time to change predictions without retraining.

  4. Quantum DeepONet gets conformal calibration: near-linear hidden layers, coverage near the 90% target

    Quantum DeepONet ensembles cut the hidden-layer cost of operator learning from the quadratic cost of dense layers to near-linear, and add a finite-sample, distribution-free lower bound on query-averaged coverage. Previously quantum operator learning returned only noisy point predictions, while Bayesian networks and Monte Carlo dropout offered no distribution-free guarantee. The approach replaces dense hidden layers with quantum orthogonal neural networks and applies split conformal calibration to the mean and standard deviation of independently trained ensembles. On synthetic antiderivative and advection benchmarks, relative L2 errors are 0.46% and 2.28%, with 8-model ensemble coverage of 92.13% and 88.99% against a 90% target.

  5. arXiv

    LLM Bayesian fine-tuning shows state-dependent heavy-tailed gradient noise: geometric-median centering plus residual clipping keeps Fisher-information guarantees on non-log-concave targets

    In Bayesian LoRA fine-tuning of LLMs and vision-language models, minibatch gradient errors are heavy-tailed and grow with distance from the final iterate, and a robust estimator combining geometric-median centering with residual clipping, together with adaptive step sizes, retains non-asymptotic Fisher-information guarantees on non-log-concave targets.

  6. arXiv

    Teacher supervision stops being copied token by token: reward-weighted distillation lifts math benchmarks by up to 6.5 points

    Distillation no longer treats every teacher token alike: each token's weight is set by its predicted effect on the student's post-update reward. Earlier token-selection methods relied on proxies such as uncertainty or teacher-student divergence, which need not identify reward-improving supervision. RAD scores each token by differentiating through a temporary optimizer step, then updates token weights by projected ascent, improving mean accuracy on six competition-math benchmarks by 6.5 points for Qwen3-1.7B and 2.5 points for Qwen3-4B-Instruct-2507 at a matched training-response budget, and reaching 54.69% and 58.67% pass@1 on LiveCodeBench-v6.

  7. arXiv

    Two-stage gated auditing cuts LLM audits from 8.15 to 2.33 per run at 72.8% detection on OpenAgentSafety

    The work presents a two-stage LLM agent monitoring framework in which single-event and trace-sequence gate rules select pending actions for inspection and an LLM audit agent judges each selected action in the context of the inspected agent's preceding trace before execution, with an optimizer LLM jointly refining gate rules and audit prompts. On the OpenAgentSafety test split, the tuned gated audit reduces mean audits from 8.15 to 2.33 per run and tokens from 47.8k to 14.6k, with 72.8% detection versus 81.5% for auditing every action and a lower false-positive rate (5.9% versus 11.8%); in two simulated multi-agent scenarios it flags all attack runs while cutting audit token usage by more than 80%.

  8. arXiv

    EDiS decomposes a graph once into cacheable edge-disjoint subgraphs and tops 19 node-classification benchmarks at matched edge budgets

    EDiS separates one-time structural extraction from per-epoch graph composition: it decomposes the graph once into cacheable edge-disjoint subgraphs using feature-based scores and successive maximum-score covering forests, then recombines them into training graphs under edge-budget constraints without re-extracting structure; across 19 homophilic, heterophilic, and large-scale node-classification benchmarks at the same edge budget against 17 baselines, EDiS-Lite achieves the highest mean benchmark score and the lowest average rank and gap-to-best.

  9. arXiv

    Purdue team's covariate-dependent CR-MRF jointly models ordinal ratings and predicts pairwise preferences more accurately than Bradley-Terry and Plackett-Luce on MovieLens

    The authors propose a covariate-dependent consecutive-ratio Markov random field (CR-MRF) that jointly models multivariate ordinal ratings, show that comparison models such as Bradley-Terry and Plackett-Luce arise as restrictions of it under independence and an exponential-race argument, and develop maximum likelihood inference with an importance-sampling estimator of the intractable normalizer; in simulations and on MovieLens, modeling ordinal scores directly yields more accurate and better-calibrated out-of-sample pairwise preference predictions than fitting comparison models to derived win-loss data.

  10. arXiv

    Purdue team's training-free one-shot segmentation framework lifts mean IoU by 33.36 points over the strongest baseline on pool-boiling images

    The work presents a training-free, one-shot framework for scientific image segmentation that uses a single annotated reference image to combine frozen DINOv3 with SAM: a background direction estimated from the reference background drives background-adaptive orthogonal projection of DINOv3 patch features to suppress artifact-related directions, cosine similarity then localizes candidate regions, and generated prompts feed SAM. Evaluated on red-blood-cell microscopy, structured-illumination pool boiling, and chest radiography, it improves mean IoU over the strongest baseline by 5.15 points on microscopy and 33.36 points on pool boiling, while matching SAM2 on chest radiographs.