All papers

Trains a reinforcement learning controller to manage frame memory dynamically for visual object tracking in foundation models.

Analyzes reconstruction failures on under-ice underwater vehicle video and introduces a tool to flag camera baseline errors.

Adapts a frozen segmentation model using a dual-branch adapter to improve cell nuclei segmentation across varied histopathology domains.

Constructs a digital twin from computed tomography scans to plan accurate surgical trajectories for ventricular shunt insertion.

Addresses cyclic preference graphs in reinforcement learning for video generation models using an anchored preference walk.

Models entity persistence and relation transitions separately to improve video scene graph generation accuracy during dynamic scene changes.

Uses representation-space supervision to initialize autoregressive video distillation students for faster and clearer generation.

Transfers capabilities from text-to-video foundation models to a unidirectional streaming model for real-time video object removal.

Combines spatial and temporal transformer modules to detect smoke and fire from camera video feeds.

Introduces dataset-level regularization stored in memory to prevent representation collapse in self-supervised learning with small batch sizes.

Learns continuous latent actions from video data using an orthogonal dictionary to build cross-domain world models.

Applies low-rank continual pretraining to vision transformers to improve task alignment in prostate magnetic resonance imaging segmentation.

Distills bidirectional video diffusion models into causal autoregressive models using an efficient single-sample supervision strategy.

Uses contrastive uncertainty learning during training to improve anatomical boundary segmentation in pelvic magnetic resonance imaging.

Introduces a geometry-grounded scene representation with continuous levels of detail to compress large view synthesis models.

Investigates the interaction between learnable image resizing and spectral-domain convolutional filter parameterization in neural networks.

Presents an adversarial evasion attack that fools face recognition systems while preserving embedded watermarks.

Compares event and conventional frame cameras via simulations for synthetic-aperture imaging through occluding scenes like foliage.

Decouples geometry and appearance learning in Gaussian reconstruction to improve three-dimensional scene rendering across diverse views.

This paper proposes a reusable, label-aware front end to allocate fixed budgets of 3D Gaussians for cinematic CT visualization without per-case optimization.

This paper presents a hybrid instance segmentation pipeline that trains ancillary models to suppress label noise when combining bounding boxes with pixel masks.

This paper assesses ten fingerprint quality assessment methods on longitudinal infant data and proposes parameter tuning routines to improve recognition.

This paper presents A3PreCore, a six-degree-of-freedom rigid-registration method for robust T1-to-FLAIR brain-mask transfer alongside quantitative quality control metrics.

This paper proposes a mixture-of-experts transformer architecture using discrete Hodge-Laplacian operators to solve steady partial differential equations on unstructured meshes.

This paper presents a feed-forward framework that encodes multi-view image features into a compact bitstream for 3D Gaussian Splatting scene reconstruction.

This paper investigates 36 handcrafted physiological markers from speech production to detect deepfake speech using standard classifiers.

The authors present IDGuardian, an adversarial perturbation framework designed to defend facial images against unauthorized personalized diffusion models.

The authors introduce DescribeEvent to show that language models can forecast temporal events when using natural-language descriptions and symbolic labels.

The paper introduces a token-level sparse attention method using a learned query-key subspace and an indexed kernel to accelerate long-context decoding.

The paper analyzes gendered reference replacements across science fiction and general fiction corpora using a masked language model.

The paper analyzes intermediate training checkpoints to trace when language models acquire morphological structures versus semantic embeddings during pretraining.

The paper examines using pre-trained language models without task-specific training for social science record linkage and demographic inference from names.

The paper shows that the Bradley-Terry model fails on Chatbot Arena for similar models, and introduces a parameter-efficient correction for ties.

The paper introduces a benchmark and similarity metric to measure the algorithmic diversity of solutions produced by code generation models.

The paper presents a four-step methodological procedure to integrate ontology-guided coding and topic modeling representations of qualitative text data.

The paper proposes a regression-based meta-evaluation framework and benchmark to assess text evaluation metrics for multilingual style-personalized generation.

The paper proposes a method combining offline execution-verified prompt updates and test-time refinement to improve program generation in language models.

The paper investigates base language model benchmarking setups to identify evaluation configurations that produce consistent, non-random, and monotonic performance signals.

The paper presents a double-annotated sentiment dataset for Nepali and evaluates several open-source and commercial language models on it.

The paper introduces a training-free sampling framework that uses internal model activations to efficiently estimate rare failure probabilities in language models.

The paper proposes a parameter-efficient method leveraging frozen model representations for multi-output text regression tasks across diverse benchmarks.

The paper presents a framework to formalize compute, labor, and workflow costs as core measurement design criteria for automated essay scoring.

The paper evaluates the ability of language model agents to diagnose appliance faults using a simulated appliance benchmark based on sensor data.

The paper presents an approach using character-level tokens, larger models, and synthetic data for the multilingual BabyLM data-efficiency challenge.

The paper introduces a benchmark to evaluate whether language models can accurately detect appropriate and inappropriate uses of persona information in dialogue.

The paper formulates task routing across language model providers as an auction mechanism to encourage truthful cost reporting while minimizing context switching.

The paper introduces a multi-stage topic modeling pipeline that merges fine-grained document clusters using a language model to balance efficiency and quality.

The paper proposes an experimental design that separates example selection from mass redistribution to analyze the effects of training-data quality filtering.

The paper shows that differences in predicted target tokens, such as whitespace, can falsely look like layer specialization across different corpora.

The paper analyzes prefix replay curricula and handoff rules under reinforcement learning to improve long-horizon task completion for language model agents.

The paper introduces a framework that decomposes user feedback into atomic rules to retain only reusable preferences for personalizing language models.

The paper proposes a multiple-summary metric and dataset to evaluate graded proposition salience across diverse text genres.

The paper introduces a reinforcement learning framework using intermediate-state branching and segment-level credit assignment for diffusion language models.

The paper presents adapted small language models designed to directly score agricultural candidate choices in a single forward pass.

The paper develops a streaming end-to-end inverse text normalization architecture using a pretrained text-to-text model with a read-tag-write policy.

The paper proposes a decoding algorithm that denoises consecutive blocks concurrently by promoting sampler states in diffusion language models.

The paper analyzes reward hacking in verifiable instruction following and introduces a likelihood bonus method to maintain answer quality.

The paper presents an item response theory framework to jointly evaluate student performance on standard questions and tutoring dialogue subtasks.

The paper introduces a threat modeling framework examining adversarial query demotion and graph edit attacks on graph-based RAG systems.

The paper presents a method using an external encoder to induce latent soft-token reasoning on frozen language model backbones.