arXiv AI
Jul 10
Validity of LLMs as data annotators: AMALIA on authority
arXiv:2607. 08731v1 Announce Type: cross Abstract: A national language model offers a linguistic community its own instrument for measuring what its citizens say and value.
Benchmarks and evaluation
Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv AI
Jul 10
Leveraging Color Naming for Image Enhancement
arXiv:2607. 08185v1 Announce Type: cross Abstract: Enhancing images to make them visually appealing is a persistent challenge in computer vision.
arXiv Machine Learning
Jul 10
Benchmark Evaluation of Feredated Learning on Multi-organ Images
arXiv:2607. 08219v1 Announce Type: cross Abstract: The privacy requirements of medical data and its substantial variations across organs and modalities hinder the clinical implementation of medical AI.
arXiv AI
Jul 10
Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks
arXiv:2607. 07907v1 Announce Type: cross Abstract: With the growing adoption of VLMs, DMs, LLMs, and AFMs, these multimodal foundation models can inadvertently encode sensitive, copyrighted, biased, or unsafe cross-modal associations that originate from their training data.
arXiv AI
Jul 10
Blind-Spots-Bench: Evaluating Blind Spots in Multimodal Models
arXiv:2607. 08317v1 Announce Type: new Abstract: Modern AI models achieve strong performance on many established benchmarks, yet they still fail on tasks that humans find almost trivial, such as manipulating a string or drawing a dog with five legs.
arXiv AI
Jul 10
When Implausible Tokens Get Reinforced: Tail-Aware Credit Calibration for LLM Reinforcement Learning
arXiv:2607. 07976v1 Announce Type: cross Abstract: Reinforcement learning (RL) has achieved remarkable success in enhancing the reasoning capabilities of large language models (LLMs).
arXiv AI
Jul 10
DreamCharacter-1: From 3D Generative Foundation Models to Product-Ready Character Generation
arXiv:2607. 07817v1 Announce Type: cross Abstract: We present DreamCharacter-1, a lightweight post-adaptation framework that calibrates pretrained 3D foundation models toward high-fidelity, production-ready 3D character generation.
arXiv AI
Jul 10
Linear Attention Architectures: Mechanisms, Trade-offs, and Cross-Layer Routing
arXiv:2607. 07953v1 Announce Type: cross Abstract: Self-attention lets each token retrieve information from the full context, but its quadratic cost in sequence length limits training and inference at long context.
arXiv AI
Jul 10
TMI: Text-to-Image Meets Image-to-Image for Complementary Data Synthesis to Boost Long-Tailed Instance Segmentation
arXiv:2607. 08201v1 Announce Type: cross Abstract: Large-vocabulary instance segmentation is constrained by long-tailed category distributions and fine-grained inter-class ambiguity.
arXiv Machine Learning
Jul 10
Predicting Viticulture Potential through an Ensemble of U-Net and a Geospatial Foundation Model
arXiv:2607. 08449v1 Announce Type: cross Abstract: Determining agricultural potential is fundamental to sustainable land management and agricultural planning.
arXiv AI
Jul 10
Open-ended Multi-agent Autocurricula via Visual Inspection of Policies with Multi-modal LLMs
arXiv:2607. 08193v1 Announce Type: cross Abstract: Open-ended curricula in Reinforcement Learning (RL) aim to train generally-capable agents by identifying tasks that facilitate learning increasingly complex skills.
arXiv Machine Learning
Jul 10
Frequency-Domain Multi-Modality Transportation Modeling
arXiv:2607. 08475v1 Announce Type: new Abstract: Multi-modality transportation refers to urban systems composed of multiple transportation modes, such as traffic flow and public transit, whose dynamics are coupled by shared temporal patterns.
arXiv Machine Learning
Jul 10
Ensemble Diversity Optimization for Subjective Supervision
arXiv:2607. 08493v1 Announce Type: new Abstract: Subjective NLP tasks often exhibit systematic annotator disagreement, requiring models that represent uncertainty rather than collapse it.
arXiv AI
Jul 10
LDFE: Laplacian Decoupled Feature Enhancement Block for Dual-Stream CNN-based RGB-IR Object Detection
arXiv:2607. 08076v1 Announce Type: cross Abstract: The complementary information between RGB and IR images can significantly enhance object detection performance under extreme conditions.
arXiv Machine Learning
Jul 10
Joint Discrete-Continuous Flow Matching for Open-Vocabulary Inverse Design of Multilayer Optical Coatings
arXiv:2607. 08392v1 Announce Type: cross Abstract: Amortized neural inverse design typically remains closed-world: component choices are fixed vocabulary tokens, coordinate grids are frozen at training time, and continuous variables are discretized into sequence tokens.
arXiv AI
Jul 10
Persona Cartography: Charting Language Model Personality Traits in Weight Space
arXiv:2607. 07916v1 Announce Type: new Abstract: Large language models exhibit recurring behavioural patterns -- personas -- that shape generalisation and safety, but we lack reliable tools for decomposing, measuring, and controlling them.
arXiv Machine Learning
Jul 10
MulTTiPop: A Multitrack Transcription Dataset for Pop Music
arXiv:2607. 08756v1 Announce Type: cross Abstract: We present MulTTiPop, a dataset of pop music segments and their associated multitrack MIDI recordings for the evaluation of automatic music transcription models.
arXiv AI
Jul 10
Infinity-Parser2 Technical Report
arXiv:2607. 07836v1 Announce Type: new Abstract: We present Infinity-Parser2, a large multimodal model that couples a controllable data-synthesis pipeline with multi-task reinforcement learning for end-to-end document parsing, addressing the persistent scarcity of faithfully annotated parsing corpora.
arXiv Machine Learning
Jul 10
Super Weights in LLMs and the Failure of Selective Training
arXiv:2607. 08733v1 Announce Type: new Abstract: Recent work identified Super Weights, individual parameters whose removal degrades model performance by orders of magnitude.
arXiv Machine Learning
Jul 10
Functional and Secure Code Generation with Task Vectors
arXiv:2607. 07881v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for code generation, but they struggle to generate functional code free of security vulnerabilities.