Knowledge catalogue
arXiv:2603.14222v2 Announce Type: replace-cross Abstract: Contrastive pretraining models such as CLIP and CLAP, serve as the ubiquitous perceptual backbones for modern multimodal large models, yet the
arXiv:2602.15353v2 Announce Type: replace-cross Abstract: Large pretrained language models and neural reasoning systems have advanced many natural language tasks, yet they remain challenged by knowled
Chroma1 is an 8.9B parameter text-to-image foundational model , with different versions including V48 (512x raw pretrained checkpoint) and variants like Chroma1-HD (mixed resolution finetune) . Chroma
arXiv:2604.19254v1 Announce Type: cross Abstract: Parameter-efficient fine-tuning (PEFT) reduces the training cost of full-parameter fine-tuning for large language models (LLMs) by training only a sma
arXiv:2506.01687v3 Announce Type: replace Abstract: Subword-level understanding is integral to numerous tasks, including understanding multi-digit numbers, spelling mistakes, abbreviations, rhyming, a
arXiv:2604.17177v1 Announce Type: new Abstract: Fine-tuning adapts pretrained networks to new objectives. Whether the resulting depth profile of representational change reflects an intrinsic property
arXiv:2604.17943v1 Announce Type: new Abstract: Open-domain RAG benchmarks over public corpora can overestimate deployment performance due to pretraining overlap and weak attribution requirements. We
arXiv:2604.18109v1 Announce Type: new Abstract: This paper presents factorized linear projection (FLiP) models for understanding pretrained sentence embedding spaces. We train FLiP models to recover t
arXiv:2604.16796v1 Announce Type: new Abstract: Generative semantic communication (SemCom) harnesses pretrained generative priors to improve the perceptual quality of wireless image transmission. Exis
arXiv:2507.09025v4 Announce Type: replace Abstract: We propose Lizard, a linearization framework that transforms pretrained Transformer-based Large Language Models (LLMs) into subquadratic architectur
arXiv:2604.18360v1 Announce Type: cross Abstract: Audio-text retrieval systems based on Contrastive Language-Audio Pretraining (CLAP) achieve strong performance on traditional benchmarks; however, the
arXiv:2604.17915v1 Announce Type: new Abstract: Vision-Language Models(VLMs) excel at autoregressive text generation, yet end-to-end autonomous driving requires multi-task learning with structured out
arXiv:2604.17257v1 Announce Type: new Abstract: Recent text embedding models are often adapted to specialized domains via contrastive pre-finetuning (PFT) on a naive collection of scattered, heterogen
arXiv:2604.16715v1 Announce Type: cross Abstract: Graph foundation models have demonstrated remarkable adaptability across diverse downstream tasks through large-scale pretraining on graphs. However,
arXiv:2511.10370v2 Announce Type: replace Abstract: Geospatial foundation models (GFMs) for Earth observation often fail to perform reliably in environments underrepresented during pretraining. We int
arXiv:2510.09378v2 Announce Type: replace Abstract: Recent efforts to accelerate LLM pretraining have focused on computationally-efficient approximations that exploit second-order structure. This rais
arXiv:2604.15678v1 Announce Type: new Abstract: Pretrained Vision-Language Models (VLMs) like CLIP show promise in continual learning, but existing Few-Shot Class-Incremental Learning (FSCIL) methods
arXiv:2510.20616v2 Announce Type: replace Abstract: Differentially private (DP) transfer learning, i.e., fine-tuning a pretrained model on private data, is the current state-of-the-art approach for tr
arXiv:2508.16860v2 Announce Type: replace-cross Abstract: Pretrained Language Models or PLMs are transformer-based architectures that can be used in bug triaging tasks. PLMs can better capture token s
arXiv:2604.07663v2 Announce Type: replace Abstract: The AdamW optimizer, while standard for LLM pretraining, is a critical memory bottleneck, consuming optimizer states equivalent to twice the model's
arXiv:2604.13508v1 Announce Type: new Abstract: Sparse Upcycling provides an efficient way to initialize a Mixture-of-Experts (MoE) model from pretrained dense weights instead of training from scratch
arXiv:2507.13942v2 Announce Type: replace Abstract: Forecasting future events is a fundamental capability for general-purpose systems that plan or act across different levels of abstraction. Yet, eval
arXiv:2604.13627v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) is a common first stage of LLM post-training, teaching the model to follow instructions and shaping its behavior as a hel
arXiv:2604.05096v2 Announce Type: replace Abstract: Large language models (LLMs) acquire most of their knowledge during pretraining, which ties them to a fixed snapshot of the world and makes adaptati
arXiv:2508.01620v3 Announce Type: replace-cross Abstract: Machine Unlearning (MU) aims to selectively erase the influence of specific data points from pretrained models. However, most existing MU meth
arXiv:2604.12277v1 Announce Type: new Abstract: Pretrained language models often rely on superficial features that appear predictive during training yet fail to generalize at test time, a phenomenon k
arXiv:2604.12820v1 Announce Type: new Abstract: Large language models (LLMs) inherently absorb harmful knowledge, misinformation, and personal data during pretraining on large-scale web corpora, with
arXiv:2604.12908v1 Announce Type: new Abstract: At its core, robotic manipulation is a problem of vision-to-geometry mapping (f(v) rightarrow G). Physical actions are fundamentally defined by geometri
arXiv:2602.08590v4 Announce Type: replace Abstract: Vision-language pretrained models offer strong transferable representations, yet adapting them in privacy-sensitive multi-party settings is challeng
arXiv:2503.10471v2 Announce Type: replace-cross Abstract: Predicting crystal structures from chemical compositions is a fundamental challenge in materials discovery, complicated by complex 3D geometri
arXiv:2604.12935v1 Announce Type: new Abstract: Efficiently merging several models fine-tuned for different tasks, but stemming from the same pretrained base model, is of great practical interest. Des
arXiv:2604.10856v1 Announce Type: cross Abstract: Open-loop (OL) to closed-loop (CL) gap (OL-CL gap) exists when OL-pretrained policies scoring high in OL evaluations fail to transfer effectively in c
arXiv:2604.10882v1 Announce Type: cross Abstract: Graph Neural Network pretraining is pivotal for leveraging unlabeled graph data. However, generalizing across heterogeneous domains remains a major ch
arXiv:2604.09694v1 Announce Type: new Abstract: Autonomous Unmanned Aerial Vehicles (UAVs) must reliably detect thin obstacles such as wires, poles, and branches to navigate safely in real-world envir
arXiv:2604.10023v1 Announce Type: cross Abstract: With the growing availability of open-sourced adapters trained on the same diffusion backbone for diverse scenes and objects, combining these pretrain
arXiv:2604.10103v1 Announce Type: new Abstract: Streaming video generation (SVG) distills a pretrained bidirectional video diffusion model into an autoregressive model equipped with sliding window att
arXiv:2604.10021v1 Announce Type: cross Abstract: Self-supervised music foundation models underperform on key detection, which requires pitch-sensitive representations. In this work, we present the fi
arXiv:2604.11508v1 Announce Type: cross Abstract: Fine-tuning pretrained image classifiers is standard practice, yet which individual samples are forgotten during this process, and whether forgetting
arXiv:2604.10123v1 Announce Type: new Abstract: Dysarthric speech severity assessment typically requires trained clinicians or supervised models built from labelled pathological speech, limiting scala
arXiv:2604.10544v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) have recently achieved remarkable success in universal forecasting by leveraging large-scale pretraining on dive
arXiv:2511.05168v2 Announce Type: replace Abstract: Vision foundation models achieve strong performance on both global and locally dense downstream tasks. Pretrained on large images, the recent DINOv3
arXiv:2604.09391v1 Announce Type: cross Abstract: Machine unlearning poses challenges in removing mislabeled, contaminated, or problematic data from a pretrained model. Current unlearning approaches a
arXiv:2604.08579v1 Announce Type: cross Abstract: We study cross-modal alignment between independently pretrained vision (DINOv2) and language (all-MiniLM-L6-v2) encoders using the functional map fram
Auto-generated index of all tools mentioned across the wiki.
arXiv:2604.06468v1 Announce Type: new Abstract: Most methods for learning with noisy labels require privileged knowledge such as noise transition matrices, clean subsets or pretrained feature extracto
arXiv:2604.08456v1 Announce Type: cross Abstract: Despite rapid progress, pretrained vision-language models still struggle when answers depend on tiny visual details or on combining clues spread acros
arXiv:2604.07941v1 Announce Type: new Abstract: Post-training has become central to turning pretrained large language models (LLMs) into aligned and deployable systems. Recent progress spans supervise
arXiv:2604.08536v1 Announce Type: new Abstract: We introduce RewardFlow, an inversion-free framework that steers pretrained diffusion and flow-matching models at inference time through multi-reward La
arXiv:2509.26036v3 Announce Type: replace Abstract: While Contrastive Language-Image Pretraining (CLIP) excels at zero-shot tasks by aligning image and text embeddings, its performance in few-shot cla
arXiv:2604.06391v1 Announce Type: cross Abstract: Graphs are a central representation in biomedical research, capturing molecular interaction networks, gene regulatory circuits, cell--cell communicati
arXiv:2604.06440v1 Announce Type: cross Abstract: Visual prompting (VP) has emerged as a popular method to repurpose pretrained vision models for adaptation to downstream tasks. Unlike conventional mo
arXiv:2610.21323v1 Announce Type: new Abstract: Most vision-language models (VLMs) treat the vision encoder as a static feature extractor: it encodes an image once, independently of the question, and
arXiv:2610.21470v1 Announce Type: new Abstract: When a vision-language model is finetuned on class-imbalanced data, rare-class recall can fall well below the model's own zero-shot performance. In this
arXiv:2610.22356v1 Announce Type: new Abstract: Vision-language models still struggle to understand 3D spatial relations across views and update spatial judgments as the observer moves. A common appro
arXiv:2610.21901v1 Announce Type: cross Abstract: Sequence models can generalise in fundamentally distinct ways: extrapolating structures seen in training, or using their prompt to adapt to novel task
arXiv:2610.21736v1 Announce Type: new Abstract: Multi-objective optimization involves diverse preferences over conflicting objectives. Rather than returning a finite set of trade-off solutions, prefer
arXiv:2610.21356v1 Announce Type: new Abstract: Healthcare institutions routinely accumulate high-dimensional computed tomography (CT) volumes, but only a fraction can be annotated by experts. We pres
arXiv:2610.20965v1 Announce Type: new Abstract: Large multimodal models have achieved remarkable progress by scaling model capacity and training on massive paired multimodal data. However, this paradi
arXiv:2610.21125v1 Announce Type: new Abstract: Few-step generative models produce high-quality samples in one or a few evaluations. However, steering them toward desired properties remains difficult,
arXiv:2610.20739v1 Announce Type: new Abstract: Quantile forecasts summarize uncertainty at each future time step, while diffusion forecasters represent it through sampled future trajectories. We stud