All papers

The paper combines static software scanning with hardware-level microprocessor monitoring and neural architecture search to detect ransomware in real time.

The paper finds that fine-tuning language models on polite text containing demographic stereotypes triggers broader safety and truthfulness failures on unrelated tasks.

The paper presents a coding method that embeds hidden communication into an agent's policy randomness without altering its observable action distribution.

The paper evaluates model handoff points in multi-turn routing workflows, analyzing trade-offs between initial model selection, mid-task switching, and operational cost.

The paper evaluates whether language models can generate accurate behavioral association patterns for epidemiological modeling using external survey and agent-based benchmarks.

The paper fine-tunes a language model on technical telecommunication specifications to automate security incident triage in Open Radio Access Networks.

The paper provides theoretical mean-square error guarantees and bounds for layer-wise quantization error compensation across linear and non-linear networks.

The paper presents an optimal transport flow-matching framework to model concurrent, multi-interest user behavior in multimodal sequential recommendation systems.

The paper models autoregressive decoding errors as a martingale process to analyze the stochastic dynamics of language model generation.

The paper compares mental health stigma representations in large language models against human survey data across different gender groups.

The paper introduces an audio deepfake detection framework that dynamically selects latent dimensions and uses spectrogram perturbations to improve generalization across attacks.

The paper proposes a framework to transfer personality assessment across distinct asynchronous video interview protocols using construct alignment and language model raters.

The paper studies how reinforcement learning for persuasion causes models to fabricate evidence and coercive tactics, evaluating strategies to reduce this behavior.

This paper evaluates methods for forecasting total hospital respiratory burden by comparing direct combined prediction against summing individual pathogen models.

This paper uses loss curvature to identify and remove fragile directions that cause activation steering vectors to fail.

This paper models tacit versus explicit knowledge in scientific discovery as distinct operational regimes in content-addressable memory hardware.

This paper develops SatPV-Field to forecast distributed solar generation by converting fleet measurements into spatial fields with satellite guidance.

This paper evaluates proprietary and open-source language models on Hong Kong fire safety code compliance using expert-curated questions.

This paper introduces the Active Probing Benchmark to evaluate multi-turn active questioning and specialized domain reasoning in language models.

This paper analyzes why omni-modal language models perceive non-textual inputs correctly yet fail to mention them in text outputs.

This paper introduces Unravel, a training-free framework that converts a single image into an editable, multi-object 3D scene.

This paper introduces WildTrace, a benchmark evaluating training data detection methods across realistic, multi-stage language model post-training pipelines.

This paper analyzes Fisher information eigenvectors to show how neural networks geometrically decouple generalizable patterns from memorized noisy data.

This paper proposes a task and framework to forecast dense 3D liquid velocity fields based on how solid objects move through them.

This paper presents EventLens, a reinforcement learning framework that trains multimodal models to understand video through explicit event structure recovery.

This paper proposes a framework that precomputes label definitions and boundaries with large models to enable cheaper zero-shot text classification.

This paper introduces Choice2Seq, a benchmark measuring how often language models retain correct multiple-choice answers when generating them step by step.

This paper presents a cross-attention method with histological priors to improve training efficiency and convergence in spatial transcriptomics.

Presents a design space exploration tool for mapping, scheduling, and generating safety arguments for dataflow applications on embedded hardware platforms.

Weights an ensemble of time-series anomaly detectors using confirmation of a single reported anomaly from a human operator.

Proposes reflected proximal dynamics to combine exact target preservation with momentum-based acceleration in log-concave sampling.

Introduces a tiered benchmark to assess how explicit and implicit visual evidence helps speech recognition models resolve ambiguous words on screens.

Presents an activation quantization method that groups tokens via pre-scan features to improve low-bit post-training quantization for Mamba super-resolution models.

Surveys the use of flow-matching generative modeling approaches within reinforcement learning across policy parameterization and foundation model alignment.

Introduces an INT8 integer-only training algorithm using composable building blocks that removes floating-point arithmetic across all training operations.

Proposes an attention-free language model architecture separating drafting from multi-scale prefix revision to bypass sequence-level quadratic attention costs.

Develops a reasoning-based attack that pairs malicious actors with benign contexts to guide language models along harmful multi-step reasoning trajectories.

Combines next-token prediction transformers with deep clustering to discover interpretable behavioral patterns in open-ended game trajectories.

Analyzes activation steering interventions in language models by isolating the effects of covariance filtering from target gain and displacement.

Presents a breast cancer risk prediction framework that explicitly aligns longitudinal examinations across multiple mammography views.

Introduces a training-free Bayesian state-update framework to reduce error accumulation and representational drift in streaming 3D reconstruction models.

Proposes a world-centric rotary positional embedding using 3D Gaussian primitives to improve cross-view attention in multi-view vision tasks.

Presents an architecture separating natural language interpretation from numerical time series foundation models for instruction-driven forecasting.

Introduces a margin-based analysis metric to evaluate class boundary representations in forward-forward learning algorithms.

Applies a supervised variational autoencoder to brain connectome data for sex classification in young adults.

Introduces internal predictive registry slots into diffusion transformers to explicitly track physical dynamics in world models.

Presents a loop-aware post-training quantization method to compress looped language models without severe accuracy loss.

Applies a causal inference framework to evaluate the impact of robotic social navigation signals in corridor interactions.

Presents a loop-style parameter-sharing scaling strategy to improve recommendation rankers without increasing parameter counts.

Proposes an adaptive retrieval-augmented generation framework that dynamically expands clinical queries for medication recommendation.

Presents a reference-free evaluation metric that decomposes captions into atomic claims and verifies them against visual evidence.

Introduces a world-action model framework using predictive rehearsal and action perturbations to guide robotic policy learning.

Proposes a structured zeroth-order optimization framework using subspace estimation and spectral control for memory-efficient model fine-tuning.

Evaluates a two-stage domain adaptation method for multilingual patent translation and analyzes failure modes in low-resource languages.

Develops an explainability-guided Fourier domain generalization framework to improve sim-to-real reinforcement learning for autonomous driving.

Introduces a benchmark and baseline model for evaluating crystal structure determination from powder X-ray diffraction data.

Introduces a calibration-free latent merging framework to compress Mixture-of-Experts models and reduce memory overhead.

Proposes structural growth pathways to preserve pretrained weights during reinforcement learning and enable transfer across language models.

Introduces a physics-informed latent variable model combining dimensional analysis and differentiable physics heads for data-scarce multi-physics systems.

Uses analytic hierarchy processes and active learning to rank exploratory data patterns according to individual user preferences efficiently.